Aggregation of [(tBu3SiS)MX]: Structures of {[Br 2Fe(μ-SSitBu3)2FeBr(THF)]Na(THF) 4}∞, cis[(THF)Fel]2(μ-SSitBu 3)2

Article abstract of DOI:10.1021/ic051289u

A convenient synthesis of ^tBu₃SiSH and ^tBu₃SiSNa(THF)_x led to the exploration of ^tBu₃SiSMX aggregation. The dimer, [( ^tBu₃SiS)Fe]₂(μ-SSi^tBU ₃)₂ (1₂), was formed from [{(Me ₃Si)₂N}Fe]₂(μ-N(SiMe₃) ₂)₂ and the thiol, and its dissolution in THF generated (^tBu₃SiS)₂Fe(THF)₂ (1-(THF) ₂). Metathetical procedures with the thiolate yielded aggregate precursors [X₂Fe](μ-SSi^tBu₃) ₂[FeX(THF)]Na(THF)₄ (3-X, X = Cl, Br) and cis-[(THF)IFe]₂(μ-SSi^tBu₃)₂ (4). Thermal desolvations of 3-Cl, 3-Br and 4 afforded molecular wheels [Fe(μ-X)(μ-SSi^tBu₃)]₁₂(C ₆H₆)_n (5-FeX, X = Cl, Br) and the ellipse [Fe(μ-l)(μ-SSi^tBu₃)]₁₄(C ₆H₆)_n (6-Fel). Related metathesis and desolvation sequences led to wheels [Co(μ-Cl)(μ-SSi^tBu ₃)]₁₂(C₆H₆)_n (5-CoCl) and [Ni(μ-Br)(μ-SSi^tBu₃)]₁₂(C ₆H₆)_n (5-NiBr). The nickel wheel disproportionated to give, in part, [(^tBu₃SiS)Ni] ₂(μ-SSi^tBu₃)₂ (7), which was also synthesized via salt metathesis. X-ray structural studies of 1₂ revealed a roughly planar Fe₂S₄ core, while 1-(THF) ₂, 3-Br, and 4 possessed simple distorted tetrahedral and edge-shared tetrahedral structures. X-ray structural studies revealed 5-MX (MX = FeCl, FeBr, CoCl, NiBr) to be wheels based on edge-shared tetrahedra, but while the pseudo-D_6d wheels of 5-FeCl, 5-CoCl, and 5-FeBr pack in a body-centered arrangement, those of pseudo-C_6v 5-NiBr exhibit hexagonal packing and two distinct trans-annular d(Br...Br). Variable-temperature magnetic susceptibility measurements were conducted on 5-FeCl, 5-CoC, 5-FeBr, and 6-Fel, and the latter three are best construed as weakly antiferromagnetic, while 5-FeCl exhibited modest ferromagnetic coupling. Features suggesting molecular magnetism are most likely affiliated with phase changes at low temperatures.

Full text of DOI:10.1021/ic051289u

Inorg. Chem. 2006, 45, 609−626
Aggregation of [(tBu3SiS)MX]: Structures of
[Br2Fe( , cis-[(THF)FeI]2(
-SSitBu3)2FeBr(THF)]Na(THF)4
[(tBu3SiS)Fe]2(
Studies of [(tBu3SiS)MX]n (M
Br, n 12; M Fe, X
{
µ
}
µ
-SSitBu3)2,
-SSitBu3)2, and Comparative Structure and Magnetism
Fe, Co, X Cl, n 12; M Fe, Ni, X
I, n 14)
∞
µ
)
)
)
)
)
)
)
)
)
Orson L. Sydora,† Thomas P. Henry, Peter T. Wolczanski,*,† Emil B. Lobkovsky,†
Evan Rumberger,‡ and David N. Hendrickson*,‡
Department of Chemistry and Chemical Biology, Baker Laboratory, Cornell UniVersity,
Ithaca, New York 14853, and Department of Chemistry and Biochemistry,
UniVersity of California, San Diego, California 92083
Received July 29, 2005
t
t
A convenient synthesis of Bu3SiSH and Bu3SiSNa(THF)x led to the exploration of “tBu3SiSMX” aggregation. The
dimer, [(tBu3SiS)Fe]2( -SSitBu3)2 (12), was formed from [
(Me3Si)2N Fe]2( -N(SiMe3)2)2 and the thiol, and its dissolution
in THF generated (tBu3SiS)2Fe(THF)2 (1
(THF)2). Metathetical procedures with the thiolate yielded aggregate
precursors [X2Fe]( X, X Cl, Br) and cis-[(THF)IFe]2(
-SSitBu3)2[FeX(THF)]Na(THF)4 (3 -SSitBu3)2 (4). Thermal
desolvations of 3 Cl, 3 Br and 4 afforded molecular wheels [Fe( -X)( FeX, X Cl, Br)
-SSitBu3)]12(C6H6)n (5
and the ellipse [Fe( -I)( FeI). Related metathesis and desolvation sequences led to wheels
-SSitBu3)]14(C6H6)n (6
[Co( -Cl)( CoCl) and [Ni( -Br)( NiBr). The nickel wheel disproportionated
-SSitBu3)]12(C6H6)n (5 -SSitBu3)]12(C6H6)n (5
to give, in part, [(tBu3SiS)Ni]2( -SSitBu3)2 (7), which was also synthesized via salt metathesis. X-ray structural
studies of 12 revealed a roughly planar Fe2S4 core, while 1 (THF)2, 3 Br, and 4 possessed simple distorted
tetrahedral and edge-shared tetrahedral structures. X-ray structural studies revealed 5 MX (MX FeCl, FeBr,
CoCl, NiBr) to be wheels based on edge-shared tetrahedra, but while the pseudo-D6d wheels of 5 FeCl, 5 CoCl,
and 5 FeBr pack in a body-centered arrangement, those of pseudo-C6v NiBr exhibit hexagonal packing and two
distinct trans-annular d(Br‚‚‚Br). Variable-temperature magnetic susceptibility measurements were conducted on
µ
{
}
µ
−
µ
−
)
µ
−
−
µ
µ
−
)
µ
µ
−
µ
µ
−
µ
µ
−
µ
−
−
−
)
−
−
−
5−
5
−FeCl, 5−CoCl, 5−FeBr, and 6−FeI, and the latter three are best construed as weakly antiferromagnetic, while
5−FeCl exhibited modest ferromagnetic coupling. Features suggesting molecular magnetism are most likely affiliated
with phase changes at low temperatures.
Introduction
for aggregation to [M(µ-X)(µ-Y)]n, which were expected to
be oligomeric or polymeric on the basis of the edge
connectivity intrinsic to tetrahedra. Literature precedent4 for
thiolate bridging groups in the development of nitrogenase-
related cluster chemistry of iron and other first-row transition
metals suggested that tBu3SiS- would be a valuable ligand.
In addition, its solubility properties would permit synthetic
studies to be carried out in nonpolar, aprotic media, and its
steric features would help ensure low coordination, as
The aggregation of low-coordinate transition metal com-
plexes can lead to a fascinating array of oligomers, polymers,
or clusters depending on the nature of the bridging ligands.
Tetrahedral species are often desirable building blocks for
extended arrays;1-3 hence, XMY equivalents were sought
* To whom correspondence should be addressed. E-mail: ptw2@
cornell.edu (P.T.W.); dhendrickson@ucsd.edu (D.N.H.).
† Cornell University.
t
investigations with the related siloxide, Bu3SiO- (silox),5
‡ University of California, San Diego.
had already shown. Herein, we report the aggregation of
(1) Zheng, C.; Hoffmann, R.; Nelson, D. R. J. Am. Chem. Soc. 1990,
112, 3784-3791.
(2) Hartl, H.; Mahdjour-Hassan-Abadi, F. Angew. Chem., Int. Ed. Engl.
1994, 33, 1841-1842.
(3) Fe´rey, G. Angew. Chem., Int. Ed. 2003, 42, 2576-2579.
(4) Dance, I. G. Polyhedron 1986, 5, 1037-1104.
10.1021/ic051289u CCC: $33.50
Published on Web 12/24/2005
© 2006 American Chemical Society
Inorganic Chemistry, Vol. 45, No. 2, 2006 609

Sydora et al.
t
[(tBu3SiS)MX] (X ) halide) to cyclic oligomers, studies of
their precursors, and some unusual magnetic properties of
these compounds. The structures of several cyclics have been
previously communicated.6,7
tBu3SiOH forms instead of the thiol. Bu3SiSH was depro-
tonated with sodium metal in THF to give the corresponding
t
sodium thiolate, Bu3SiSNa(THF)x, in 92% yield (eq 7).
t
Overall, the two step synthesis yields Bu3SiSH in 6 d with
t
a 66% yield from BuSiH (addition of triflic acid yields
tBu3SiOTf (78%)),11 while the original multistep route takes
Results
t
8 d and provides thiol in 34% yield from Bu3SiH.
1. Thiolate Synthesis. The original multistep synthesis
of tBu3SiSH (eq 1-4), developed separately by Wiberg8 and
Henry,9 consisted of the sodium metal reduction of tBu3SiBr
in THF to afford the supersilanide tBu3SiNa(THF)210 in 98%
yield followed by double nucleophilic substitution of sulfur
monochloride to produce tBu3Si-SS-SitBu3 (74%). Subse-
quent reduction of the disulfide with 0.9% Na/Hg in THF
2. The Dimer, [(tBu3SiS)Fe]2(µ-SSitBu3)2. 2.1. Synthesis
and Reactivity. Treatment of [(Me3Si)2N]2Fe12 with 4 equiv
of Bu3SiSH afforded orange crystalline [(tBu3SiS)Fe]2-
t
(µ-SSitBu3)2 (12) in 86% yield according to eq 8. H NMR
1
spectra of 12 in benzene-d6 revealed two broad resonances
at
t
generated the thiolate Bu3SiSNa(THF)x (x ) 1.4-1.9), but
C6H6
2[(Me3Si)2N]2Fe + 4tBu3SiSH 23 °C, 30 min8
with inconsistent yields (∼55%). The sodium thiolate was
protonated with anhydrous HCl and purified by sublimation,
[(tBu3SiS)Fe]2(µ-SSitBu )
t
3 2 + 4(Me3Si)2NH (8)
yielding the thiol Bu3SiSH (84%). LiAlH4 proved to be a
12
superior reducing agent, which, upon acidic workup and
sublimation, afforded the thiol in 83% yield (eq 5).
THF
12 23°C8
2(tBu SiS) Fe(THF)
2
THF
(9)
3
2
tBu3SiBr + 2 Na0
8 tBu3SiNa(THF)2 + NaBr (1)
∆, 12 h
1-(THF)2
THF
2tBu3SiNa(THF)2 + Cl-SS-Cl -78 °C8
hexane/Me3SiOSiMe3
t
2
8
( Bu3SiS)3Fe(THF)
tBu3Si-SS-SitBu3 + 2NaCl (2)
23 °C, 3 wk
2-THF
12 + tBu3Si-SS-SitBu3 + 2THF (10)
THF
tBu3Si-SS-SitBu3 + 2Na/Hg
8 2tBu3SiSNa(THF)x
23 °C, 3 d
THF
(3)
FeCl3(THF) + 3tBu3SiSNa(THF)x 23 °C, 6 h8
t
THF
23 °C, 30 min
tBu3SiSNa(THF)x + HCl
8 tBu3SiSH + NaCl (4)
+ 3 NaCl (11)
( Bu3SiS)3Fe(THF)
2-THF
1) THF, ∆, 2 d
tBu3Si-SS-SitBu3 + LiAlH4 (excess)
8
δ 1.91 (ν
≈ 200 Hz) and 2.63 (ν1/2 ≈ 200 Hz) that
1/2
2) H2O, HCl (aq)
correspond to the different thiolates. Addition of 2 equiv of
tBu3SiSNa to 12 failed to elicit the desired tris-[(tBu3SiS)3Fe]-
Na(THF)x complex13 and led to decomposition. In THF, 12
was cleaved to afford (tBu3SiS)2Fe(THF)2 (1-(THF)2, 46%
2tBu3SiSH (5)
This rather tedious sequence was shortened considerably
t
when Wiberg reported that the triflate, Bu3SiOTf,11 could
1
isolated yield, eq 9), whose H NMR spectrum showed
be substituted with anhydrous sodium hydrogensulfide upon
reluxing in THF (eq 6), whereas substitution of tBu3SiBr did
not occur. Care must be taken to use rigorously anhydrous
NaSH or
resonances attributable to the thiolate δ 2.84 (ν1/2 ≈ 40 Hz)
and THF (δ 1.87 (ν1/2 ≈ 30 Hz), 4.68 (ν1/2 ≈ 80 Hz)) ligands.
Dimer 12 could be generated concomitant with disulfide from
(tBu3SiS)3Fe(THF) (2-THF) in hexane/Me3SiOSiMe3 after
a 3 week period at 23 °C (eq 10). The purple ferric thiolate
tBu3SiOTf + NaSH THF 8 tBu3SiSH + NaOTf
(6)
t
∆, 5 d
2-THF was prepared from FeCl3 and Bu3SiSNa in 85%
yield (eq 11) and exhibited a single broad resonance in its
1H NMR spectrum at δ 21.4 (ν1/2 ≈ 1100 Hz). It degraded
to 1-(THF)2 over the course of a week in THF but much
more rapidly if heated or photolyzed.
2.2. Structure of [(tBu3SiS)Fe]2(µ-SSitBu3)2 (12). Table
1 gives the appropriate data acquisition and refinement
parameters for the X-ray structure determination of the dimer,
[(tBu3SiS)Fe]2(µ-SSitBu3)2 (12), and Table 2 gives some
pertinent interatomic distances and angles. As Figure 1
reveals, 12 has essentially a planar Fe2(µ-S)2 diamond
THF
tBu3SiSH + Na° (excess)
8 tBu3SiSNa(THF)x + 1/2H2
∆, 12 h
(7)
(5) (a) Sydora, O. L.; Wolczanski, P. T.; Lobkovsky, E. B.; Buda, C.;
Cundari, T. R. Inorg. Chem. 2005, 44, 2606-2618. (b) Veige, A. S.;
Slaughter, L. M.; Lobkovsky, E. B.; Wolczanski, P. T.; Matsunaga,
N.; Decker, S. A.; Cundari, T. R. Inorg. Chem. 2003, 42, 6204-6224.
(6) Sydora, O. L.; Wolczanski, P. T.; Lobkovsky, E. B. Angew. Chem.,
Int. Ed. 2003, 42, 2685-2687.
(7) Sydora, O. L.; Wolczanski, P. T.; Lobkovsky, E. B.; Rumberger, E.;
Hendrickson, D. N. Chem. Commun. 2004, 650-651.
(8) Wiberg, N. Coord. Chem. ReV. 1997, 163, 217-252.
(9) Henry, T. P.; Wolczanski, P. T., unpublished results.
(10) Wiberg, N.; Amelunxen, K.; Lerner, H.-W.; Schuster, H.; No¨th, H.;
Krossing, I.; Schmidt-Amelunxen, M.; Seifert, T. J. Organomet. Chem.
1997, 542, 1-18.
(12) Andersen, R. A.; Faegri, K. Jr.; Green, J. C.; Haaland, A.; Lappert,
M. F.; Leung, W.-P.; Rypal, K. Inorg. Chem. 1988, 27, 1782-1786.
(13) MacDonnell, F. M.; Ruhlandt-Senge, K.; Ellison, J. J.; Holm, R. H.;
Power, P. P. Inorg. Chem. 1995, 34, 1815-1822.
(11) Wiberg, N.; Schuster, H. Chem. Ber. 1991, 124, 93-95.
610 Inorganic Chemistry, Vol. 45, No. 2, 2006

Aggregation of [(tBu3SiS)MX]
Table 1. Crystallographic Data for [(tBu3SiS)Fe]2(µ-SSitBu3)2 (12), (tBu3SiS)2Fe(THF)2 (1-(THF)2, [Br2Fe](µ-SSitBu3)2[FeBr(THF)]Na(THF)4 (3-Br)
12
1-(THF)2
3-Br
4
formula
fw
C24H54FeS2Si2
518.84
C32H70FeO2S2Si2
663.05
C44H94O5NaSi2S2Br3Fe2
1197.93
C32H70O2Si2S2I2Fe2
972.70
space group
Z
Pbcn
8
P1h
2
P21/c
4
Pna21
4
a, Å
b, Å
c, Å
R, deg
â, deg
γ, deg
V, Å3
19.382(4)
17.872(4)
17.842(4)
90
90
90
6180.4(23)
1.115
0.710
12.897(2)
13.2170(10
13.2200(10)
98.100
104.320(10)
109.860(10)
1990.1(4)
1.106
24.363(2)
13.3296(10)
17.8294(10)
90
90.000(2)
90
5790.0(8)
1.374
2.729
31.884(6)
9.0048(16)
15.318(3)
90
90
90
4398.0(14)
1.469
2.237
F
calc, g‚cm-3
µ, mm-1
0.568
293(2)
temp, K
293(2)
173(2)
173(2)
λ (Å)
0.71073
R1 ) 0.0633
wR2 ) 0.1602
0.71073
R1 ) 0.0494
wR2 ) 0.1311
0.71073
0.71073
R1 ) 0.0460
wR2 ) 0.1202
R1 ) 0.0479
wR2 ) 0.1211
1.167
R ind [I > 2σ(I)]a,b
R1 ) 0.1103
wR2 ) 0.2532
R1 ) 0.1109
wR2 ) 0.2572
1.197
R ind (all data)a,b
GOFc
1.056
0.991
2 1/2
a R1 ) ∑||Fo| - |Fc||/∑|Fo|. b wR2 ) [∑w(|Fo| - |Fc|)2/∑wFo ]
p ) number of parameters.
.
c GOF (all data) ) [∑w(|Fo| - |Fc|)2/(n - p)]1/2, n ) number of independent reflections,
Figure 1. Molecular views of [(tBu3SiS)Fe]2(µ-SSitBu3)2 (12): top (a) and side (b).
core,13-15 with 2.323(2) and 2.333(2) Å iron-sulfur dis-
tances, a d(Fe‚‚‚Fe) of 2.982(2) Å, and Fe-S-Fe and
S-Fe-S angles of 79.64(7)° and 100.11(7)°, respectively.
The terminal thiolate possesses a slightly shorter iron-sulfur
distance of 2.218(2) Å, and its sulfur is positioned slightly
“below” the core (S1-Fe-S2/2a ) 128.65(8)°/129.62°)
is given in Table 1. Table 2 lists selected metric parameters,
and the complex is illustrated in Figure 2. 1-(THF)2 possesses
2.2711(12) and 2.2885(11) Å iron-sulfur distances, and
2.094(3) and 2.136(3) Å iron-oxygen bond lengths. The
thiolates are significantly splayed away from each other, as
the S1-Fe-S2 angle of 137.33(5)° attests, while the
O-Fe-O angle is 86.83(12)°, and the remaining O-Fe-S
angles are 100.72(9)°, 102.74(8)°, 104.53(8)° and 113.00-
(9)°. Overall the molecule is a distorted tetrahedral species,
t
while the corresponding Bu3Si group is angled (Fe-S1-
t
Si1 ) 115.79(9)°) further below. The Bu3Si group of the
bridging thiolate is positioned “above” the core (Fe1/1a-
S2-Si2 ) 123.80(10)°/127.37(10)°, as steric influences
dictate C2V symmetry.
2.3. Structure of (tBu3SiS)2Fe(THF)2 (1-(THF)2). Dif-
ficulties in assigning a structure based on broad NMR spectra
led to the X-ray structural characterization of (tBu3SiS)2Fe-
(THF)2 (1-(THF)2), whose data and refinement information
t
with the Bu3Si groups twisted away from one another
(Fe-S-Si ) 121.07(8)°, 125.33(6)°) to give the molecule
approximate C2 symmetry.
3. Cyclic Oligomer Precursors. 3.1. [X2Fe](µ-SSitBu3)2-
[FeX(THF)]Na(THF)4 (3-X, X ) Cl, Br). Since the
t
presence of two Bu3SiS groups was enough to hamper
(14) Power, P. P.; Shoner, S. C. Angew. Chem., Int. Ed. Engl. 1991, 30,
oligomerization, one thiolate was replaced in order to foment
aggregation. A mixture of FeBr2(THF)2 and Bu3SiSNa-
(THF)x was stirred in THF for 18-24 h at 23 °C to afford
330-332.
t
(15) Komuro, T.; Kawaguchi, H.; Tatsumi, K. Inorg. Chem. 2002, 41,
5083-5090.
Inorganic Chemistry, Vol. 45, No. 2, 2006 611

Sydora et al.
Table 2. Selected Interatomic Distances (Å) and Angles (deg) for
[(tBu3SiS)Fe]2(µ-SSitBu3)2 (12), (tBu3SiS)2Fe(THF)2 (1-(THF)2),
[Br2Fe](µ-SSitBu3)2[FeBr(THF)]Na(THF)4 (3-Br), and
cis-[(THF)IFe]2(µ-SSitBu3)2 (4)
12
1-(THF)2
3-Br
4
Fe‚‚‚Fe
Fe-Sb
2.982(2)
2.323(2)
3.460(4)
3.374(3)
2.382(3),a,b
2.387(3)a,c
2.348(3),b,d
2.347(3)c,d
2.358(3),a,b
2.356(3)a,c
2.374(3),b,d
2.351(3)c,d
2.333(2)
2.218(2)
Fe-St
Fe-X
2.271(2)
2.289(2)
2.403(2),a,e
2.408(2)a,f
2.379(2)d
2.5717(12)
2.5933(13)
2.062(6)a
2.011(5)d
Fe-O
2.094(3) 2.026(7)d
2.136(3)
Na-Br
3.011(4),e
3.232(4)f
Figure 2. Molecular view of (tBu3SiS)2Fe(THF)2 (1-(THF)2).
Na-O
Fe-S-Fe
2.35(4)ave
79.64(7)
93.9(1),b
90.96(10),b
91.58(11)c
94.1(1)c
analysis in view of the evident desolvation was problematic,
and 1H NMR spectra in THF-d8 revealed only a single broad
feature at δ 12.46 (ν1/2 ≈ 280 Hz); hence, the formula shown
was obtained from the X-ray crystallographic study below.
With this structure in hand, the elemental analysis may be
interpreted as corresponding to a partially desolvated 3-Br
(i.e., [Br2Fe](µ-SSitBu3)2[FeBr(THF)]Na(THF)). Since the
compound is a polymer in the solid state, it is likely that the
spectra correspond to the [Br2Fe](µ-SSitBu3)2[FeBr(THF)]-
anion. Evans’ method16 investigations were consistent with
high-spin Fe(II) centers, as a µeff/Fe of 4.5 µB was deter-
mined.
Sb-Fe-Sb
100.11(7)
85.2(1),a
86.8(1)d
87.24(8),a
86.98(8)d
St-Fe-St
Sb-Fe-St
137.33(5)
128.65(8)
129.62(8)
S-Fe-X
113.7(2),a,b,e 122.34(10),a,b
112.0(2)a,b,f 123.99(9)a,c
113.8(2),a,c,e 121.58(9),b,d
111.0(2)a,c,f
125.9(2),b,d
123.0(2)c,d
104.8(2)b,d
104.3(2)c,d
124.95(9)c,d
S-Fe-O
100.72(9)
102.74(8)
104.53(8)
113.00(9)
105.5(2)a,b
105.6(2)a,c
108.2(2)b,d
107.9(2)c,d
t
A mixture of FeCl2(THF)2 and 1 equiv of Bu3SiSNa-
X-Fe-X
O-Fe-O
X-Fe-O
116.99(7)
108.5(2)
(THF)x produced yellow crystals of [Cl2Fe](µ-SSitBu3)2-
[FeCl(THF)]Na(THF)4 (3-Cl, 74%), which was assigned its
formula on the basis of characteristics related to 3-Br.
Degradation of 3-Cl with DCl/D2O in D3COD provided the
same 1:3.9 thiolate/THF ratio, and desolvation was also
noted. Elemental analysis on one particular sample was
consistent with a desolvated version, i.e., [Cl2Fe](µ-SSitBu3)2-
[FeCl(THF)]Na. 1H NMR spectroscopy in THF-d8 revealed
a broad resonance at δ 9.11 (ν1/2 ≈ 940 Hz), and Evans’
method16 studies gave a µeff/Fe of 4.8 µB, again consistent
with high-spin ferrous centers.
86.83(12)
109.2(2),a
105.6(2)d
Fe-St-Si
Fe-Sb-Si
115.79(9)
121.07(8)
125.33(6)
123.80(10)
127.37(10°)
133.9(2),a,b
134.3(2)a,c
131.2(2),c,d
131.3(2)b,d
158.26(10),e
170.83(10)f
174.5(2)e,f
90.0(43)ave
172.6(4),
132.6(2),a,b
132.0(2)a,c
133.3(2),b,d
133.4(2)c,d
Fe-Br-Na
Br-Na-Br
O-Na-O
179.8(4)
90.0(46)ave
3.2. Structure of [Br2Fe](µ-SSitBu3)2[FeBr(THF)]Na-
(THF)4 (3-Br). Crystal and refinement data are given in
Table 1, and pertinent interatomic distances and angles are
provided in Table 2. Although the data for 3-Br is marginal,
perhaps due to twinning problems inferred from broad
reflections, the critical metric parameters are reasonable
O-Na-Br
a Fe(1) distances. b Distances and angles to S(1). c Distances and angles
to S(2). d Fe(2) distances. e Distances and angles to Br(3). f Distances and
angles to Br(5).
yellow crystals of [Br2Fe](µ-SSitBu3)2[FeBr(THF)]Na(THF)4
(3-Br, 86%) upon slow evaporation (eq 12). Degradation
of a fresh sample of 3-Br with a solution of DCl/D2O in
t
despite a severe disorder of the Bu groups. Somewhat
surprisingly, as Figure 3 indicates, 3-Br possesses a
polymeric structure in the solid state, as the square planar
Na(THF)4 unit is complexed by axial µ-Br ligands derived
from the FeBr2 side of the anion to form a zigzag chain.
Bond distances and angles of the pseudo-tetrahedral Fe(II)
centers were normal, and the iron-iron distance of
3.460(4) Å precludes significant through-space interaction.17
THF, -NaX
2FeX2(THF)2 + 2tBu3SiSNa(THF)x 18-24 h, 23 °C8
[X2Fe](µ-SSitBu3)2[FeX(THF)]Na(THF)4
(12)
3-X, X ) Cl, Br
D3COD revealed a thiolate/THF ratio of 1:3.9, as determined
by 1H NMR spectroscopy. Visible desolvation of the crystals
was noted when they remained at 23 °C for a few hours or
when exposed to vacuum. Interpretation of the elemental
(16) (a) Evans, D. F. J. Chem. Soc. 1959, 2003-2005. (b) Sur, S. K. J.
Magn. Reson. 1989, 82, 169-173. (c) Schubert, E. M. J. Chem. Educ.
1992, 69, 62.
612 Inorganic Chemistry, Vol. 45, No. 2, 2006

Aggregation of [(tBu3SiS)MX]
Figure 3. Molecular views of [Br2Fe](µ-SSitBu3)2[FeBr(THF)]Na(THF)4 (3-Br): (a) a “monomer” unit, consisting of a [Br2Fe](µ-SSitBu3)2[FeBr(THF)]-
anion paired with a Na(THF)4 cation, and (b) the zigzag chain showing cations bound to the FeBr2 end of the anion.
The d(Fe-µ-S) distances are slightly longer (2.382(3),
2.387(3) Å vs 2.348(3), 2.347(3) Å) to the iron (Fe1)
containing the Na-associated bromides, whose Fe1-(µ-Br)
distances (2.403(2), 2.408(2) Å) are slightly longer than the
Fe2-Br bond length of 2.379(2) Å. Presumably, this is a
consequence of greater anionic character in the Br2Fe(1)-
(µ-SR)2 side. The iron-oxygen distance is 2.026(7) Å, which
is significantly shorter than that in 1-(THF)2. The ligation
of the bromides to Na+ is nearly linear (i.e., Br-Na-Br )
174.5(2)°), while asymmetric Na-Br interactions are noted
(Br3, 3.011(4); Br5, 3.232(4) Å), and the chain is somewhat
kinked from iron to iron (Fe1-Br3-Na ) 158.26(10)°,
Fe1-Br5-Na ) 170.83(10)°). The diamond core is slightly
splayed at sulfur (Fe-S-Fe ) 93.9(1)°, 94.1(1)°; S-Fe-S
) 85.2(1)°, 86.8(1)°), and the polymer end of the dinuclear
anion has large Br3-Fe1-Br5 (117.0(1)°) and S-Fe1-Br
(112.6(14)°ave) angles. The remaining side of the core
manifests opened S-Fe2-Br angles (124.5(21)°ave) and
smaller ones involving the THF oxygen: O6-Fe2-S1/S2
) 104.8(2)°, 104.3(2)° and Br1-Fe2-O6 ) 108.5(2)°.
3.3. cis-[(THF)IFe]2(µ-SSitBu3)2 (4). Unlike the chloride
and bromide, when FeI2(THF)2 was treated with 1 equiv of
tBu3SiSNa(THF)x in THF, no white material precipitated
from the yellow solution. Solvent removal, thermolysis, and
crystallization from benzene afforded yellow crystalline 4
in 75% yield (eq 13). It seems plausible that a dianionic
dimer such as
1) THF
2FeI2(THF)2 + 2tBu3SiSNa(THF)x 24 h, 23 °C8 2) -THF, ∆8
3) C6H6
cis-[(THF)IFe]2(µ-SSitBu3)2
(13)
4
[{I2Fe}2(µ-SSitBu3)2][Na(THF)x]2 formed initially, and only
upon removal of solvent did loss of NaI occur. Dimer 4 did
not exhibit significant desolvation and DCl/D2O degradation
1
in CD3OD led to a thiolate/THF ratio of 1:1 by H NMR
analysis. The magnetic moment at room temperature was
4.8 µB/Fe as expected for two noninteracting ferrous centers.16
3.4. Structure of cis-[(THF)IFe]2(µ-SSitBu3)2 (4). Table
1 contains crystal and data collection information for 4, while
Table 2 provides some interatomic distances and angles. As
Figure 4 illustrates, 4 has the expected edge-shared ditetra-
hedral geometry with bridging thiolates but possesses C2V
symmetry. Non-centrosymmetric dimers without chelating
groups are unusual, but it is unclear why. Since both plausible
geometries (i.e., cis, C2V and trans C2h) possess the same
(17) Cotton, F. A.; Walton, R. A. Multiple Bonds Between Metal Atoms.
Oxford University Press: New York, 1993.
Inorganic Chemistry, Vol. 45, No. 2, 2006 613

Sydora et al.
resonance was observable at δ 4.84 (υ1/2 ≈ 60 Hz). It is
uncertain whether this signal corresponds to 5-FeBr or a
small impurity or decomposition product. Elemental analysis
was consistent with the empirical formula [FeBr(SSitBu3)],
but the low solubility precluded molecular weight studies;
hence, a single-crystal X-ray structure determination proved
necessary.
Thermal desolvation of solid 3-Cl at 80 °C under vacuum
for 1.5 h followed by benzene extraction and filtration gave
yellow crystalline 5-FeCl in moderate yield (42%, eq 14).
1
Degradation by DCl/D2O in CD3OD permitted H NMR
analysis of the thiolate/THF/C6H6 ratio to be 1.0:0.0:0.6,
which again indicated complete removal of THF. Elemental
analysis was roughly consistent with the empirical formula
[FeCl(SSitBu3)], but for 5-FeCl, no resonance was observ-
able in C6D6 solution due to insolubility. Since its solubility
appeared distinct from 5-FeBr and the thiolate/benzene
degradation ratio was different, a single-crystal X-ray
structure determination was conducted on 5-FeCl.
4.2. [Fe(µ-I)(µ-SSitBu3)]14(C6H6)n (6-FeI). Solid 4 was
heated at 117 °C under vacuum for 5 h and extracted into
benzene. Dark yellow crystals of 6-FeI were isolated in poor
yield (16%) upon slow evaporation. The formulation is
Figure 4. Molecular view of cis-[(THF)IFe]2(µ-SSitBu3)2 (4).
rotational symmetry,18 they must be entropically similar, and
there is no obvious enthalpy advantage for either. Perhaps
intermolecular dipole-dipole interactions provide a favorable
situation, since 4 crystallizes in the centrosymmetric ortho-
rhombic space group Pna21.
1) 117 °C, vacuum, 5 h, -THF8
cis-[(THF)IFe]2(µ-SSitBu3)2
2) benzene
4
[Fe(µ-I)(µ-SSitBu3)]14(C6H6)n
6-FeI
The Fe-Fe interatomic distance is 3.374(3) Å, which
precludes any significant Fe-Fe interaction,17 and the bond
lengths and angles in the dimer are normal. The d(FeI) are
2.5933(13) and 2.5717(12) Å, the iron-sulfur distances
average 2.360(3) Å, and the d(Fe-O) are 2.062(6) and
2.011(5) Å, all values consistent with a ferrous iron. As
the S-Fe-S (86.98(8)° and 87.24(8)°) and Fe-S-Fe
(90.96(10)° and 91.58(11)°) angles show, the Fe2(µ-S)2
diamond core is slightly puckered in response to steric
(15)
based on its degradation by DCl/D2O in D3COD, which
established a thiolate/THF/benzene ratio of 1.0:0.0:0.9 ratio
according to 1H NMR spectral analysis. Elemental analysis
failed despite submission of well-defined, large crystals.
4.3. [Co(µ-Cl)(µ-SSitBu3)]12(C6H6)6 (5-CoCl). Since a
practical recipe for aggregation was discovered, the methods
were extended to the remaining first row “group VIII” metals.
t
interactions between THF and the Bu3SiS groups. The
t
O-Fe-I angles are 109.19(15)° and 105.57(15)°, while the
I-Fe-S angles (123.2(15)°ave) are splayed to a far greater
extent than the O-Fe-S angles (106.8(14)°ave).
A THF mixture of anhydrous CoCl2 and Bu3SiSNa(THF)x
was stirred for 12 h to provide a blue solution. The THF
was removed and the resulting blue material, conceivably
the Co version of 3-Cl or 4, turned emerald green upon
thermolysis under vacuum. Benzene extraction and filtration
afforded green crystals of
t
4. Cyclic Oligomer Synthesis. 4.1. [Fe(µ-X)(µ-SSiBu3)]12-
(C6H6)n (5-FeX, X ) Cl, Br). Solid 3-Br was heated at
79°C under vacuum for 2 h and extracted with benzene to
produce yellow crystals of 5-FeBr in 72% yield (eq 14)
after filtration. Degradation of the crystals with a solution
of DCl/D2O in D3COD revealed a thiolate/THF/C6H6 ratio
1) THF, 23 °C, 12 h
2) 81°C, 1.5 h vacuum,
-THF, -NaX
CoCl2 + tBu3SiSNa(THF)x
8
1
of 1.0:0.0:1.0, as determined by H NMR spectroscopic
3) benzene
[Co(µ-Cl)(µ-SSitBu3)]12(C6H6)6
5-CoCl
integration. Complete
(16)
1) 80°C, vacuum,
1.5-2 h, -THF, -NaX8
[X2Fe](µ-SSitBu3)2[FeX(THF)]Na(THF)4
3-X, X ) Cl, Br
5-CoCl in modest yield (21%, eq 16). Degradation of
5-CoCl by DCl/D2O in D3COD revealed a thiolate/THF/
C6H6 ratio of 1.0:0.0:0.5. Elemental analysis was consistent
with the empirical formula [CoCl(SSitBu3)], and its structure
was determined by single-crystal X-ray diffraction methods.
4.4. [Ni(µ-Br)(µ-SSitBu3)]12(solvent)n (5-NiBr). Synthe-
sis of a related nickelous cyclic oligomer was not straight-
2) benzene
[Fe(µ-X)(µ-SSitBu3)]12(C6H6)n
(14)
5-FeX, X ) Cl, Br
THF desolvation had occurred. The remaining material
proved to be only partially soluble in C6D6, and a broad
t
forward. The combination of NiBr2(THF)2 and Bu3SiSNa-
(THF)x in THF produced a green solution, but thermal
(18) Lowry, T. H.; Richardson, K. S. Mechanism and Theory in Organic
Chemistry, 3rd ed.; Harper and Row: New York, 1987.
614 Inorganic Chemistry, Vol. 45, No. 2, 2006

Aggregation of [(tBu3SiS)MX]
Table 3. Crystallographic Data for [Fe(µ-Cl)(µ-SSitBu3)]12(C6H6)7 (5-FeCl), [Co(µ-Cl)(µ-SSitBu3)]12(C6H6)6 (5-CoCl), [Fe(µ-Br)(µ-SSitBu3)]12
(5-FeBr), [Ni(µ-Br)(µ-SSitBu3)]12 (5-NiBr), and [Fe(µ-I)(µ-SSitBu3)]14 (6-FeI)a
5-FeCl
5-CoCl
5-FeBr
5-NiBr
6-FeI
formula
fw
space group
Z
C
46.5H91.5Si3S3Cl3Fe3
C45H90Si3S3Cl3Co3
1094.79
P4h21c
C36H81Si3S3Fe3Br3
1101.74
P4h21c
C24H54Si2S2Ni2Br2
740.24
P63mc
C50H99.5Si3.5S3.5Fe3.5I3.5
1550.99
Cmca
1105.08
P4h21c
8
8
8
12
16
a, Å
b, Å
c, Å
R, deg
â, deg
γ, deg
V, Å3
23.3716(7)
23.3716(7)
24.0912(11)
90
90
90
13159.4(8)
1.116
0.950
23.287(3)
23.287(3)
24.155(4)
90
90
90
13098(3)
1.110
1.048
24.542(5)
24.542(5)
24.086(8)
90
90
90
13349(6)
1.096
2.605
26.128(16)
26.128(16)
24.69(2)
90
90
120
14597(18)
1.010
2.559
38.173(6)
24.296(4)
35.977(5)
90
90
90
33367(9)
1.235
2.055
F
calc, g‚cm-3
µ, mm-1
temp, K
λ (Å)
173(2)
173(2)
173(2)
173(2)
173(2)
0.71073
R1 ) 0.0432
wR2 ) 0.1080
R1 ) 0.0612
wR2 ) 0.1116
1.075
0.71073
R1 ) 0.0533
wR2 ) 0.1255
R1 ) 0.0717
wR2 ) 0.1338
1.013
0.71073
R1 ) 0.0615
wR2 ) 0.1414
R1 ) 0.0895
wR2 ) 0.1592
0.793
0.71073
R1 ) 0.1491
wR2 ) 0.3646
R1 ) 0.2261
wR2 ) 0.4001
1.128
0.71073
R1 ) 0.0762
wR2 ) 0.1952
R1 ) 0.1272
wR2 ) 0.2390
1.067
R ind
[I > 2σ(I)]b,c
R ind
(all data)b,c
GOFd
a For explanations regarding the variable amounts of benzene of solvation, see the text and Experimental Section. b R1 ) ∑||Fo| - |Fc||/∑|Fo|. c wR2 )
2
[∑w(|Fo| - |Fc|)2/∑wFo ]1/2
.
d GOF (all data) ) [∑w(|Fo| - |Fc|)2/(n - p)]1/2, n ) number of independent reflections, p ) number of parameters.
desolvation and benzene extraction of the resulting green
solid led to a complex mixture in which red 5-NiBr and
purple [(tBu3SiS)Ni]2(µ-SSitBu3)2 (7, eq17)
as well. Thermolysis of the green solid in eq 19 led to a
red-purple solid, but the purple dimer (7) could be washed
away in Et2O, and extraction in hexane led to a red solution
from which 5-NiBr crystallized in ∼15% yield. It was
typically contaminated with small amounts of 7 and a white
solid,
1) THF, 23°C, 16 h
2) 88 °C,
NiBr2(THF)2 + tBu3SiSNa(THF)x 3 h vacuum, -THF, -NaX8
3) benzene or hexane
hexane8
[Ni(µ-Br)(µ-SSitBu3)]12(C6H6)n (5-NiBr)
[Ni(µ-Br)(µ-SSitBu3)]12
5-NiBr
(17)
12 h
+ [(tBu3SiS)Ni]2(µ-SSitBu3)2 (7) + ‚‚‚
t
t
NiBr (s) +
(20)
[( Bu3SiS)Ni]2(µ-SSi Bu3)2
2
THF
NiBr2(THF)2 + 2tBu3SiSNa(THF)x 24 h, - 2NaBr8
7
[(tBu3SiS)Ni]2(µ-SSitBu3)2
(18)
presumably NiBr2, thereby precluding elemental analysis and
bulk magnetic studies. If the red solution was allowed to
stand overnight, it turned purple, indicative of the presence
of 7 (eq 20), apparently formed via disproportionation.
5. Cyclic Oligomer Structures. X-ray structure determi-
nations of the cyclic oligomers [M(µ-X)(µ-SSitBu3)]n (M )
Fe, n ) 12, X ) Cl (5-FeCl), Br (5-FeBr); M ) Fe, n )
14, X ) I (6-FeI);6 M ) Co, n ) 12, X ) Cl (5-CoCl);
M ) Ni, n ) 12, X ) Br (5-NiBr))7 have been previously
communicated, but some comparison of the compounds is
warranted. Table 3 provides an abbreviated list of the
crystallographic data for the 12-membered wheels (5-MX)
and the 14-membered ellipse (6-FeI), while Table 4 provides
salient interatomic distances and angles. Note that there is a
discrepancy between the number of solvent molecules
implicated by the degradation studies in the previous section
and those inferred by the formulas in Table 3, which are
obtained from analysis of the diffraction data. The formulas
and quality of the structures reflects the data quality and the
disorder evident in the tBu3Si groups and solvent molecules.
The quality is roughly 5-FeCl g 5-CoCl > 5-FeBr >
6-FeI . 5-NiBr. The ferrous (5-FeCl) and cobaltous (5-
CoCl) chloride wheels were well-ordered, and 2.5 benzenes
of solvation in the asymmetric unit could be modeled with
partial occupancies. Figure 5 illustrates pseudo-D6d 5-FeCl,
7
were ultimately identified. Diamagnetic, purple crystalline
7 was synthesized from NiBr2(THF)2 and 2 equiv of the
thiolate in 63% yield (eq 18). 1H and 13C{1H} NMR spectra
of 7 revealed two Bu3Si moieties, and molecular weight
studies by the Signer isopiestic method were indicative of a
t
dimer.13-15
The reaction of NiBr2(THF)2 and 1 equiv of thiolate was
assayed prior to desolvation by removing THF and crystal-
lizing a green product, tentatively formulated as as [(THF)-
BrNi]2(µ-SSitBu3)2 (8, eq 19).
THF, 23 °C
NiBr2(THF)2 + tBu3SiSNa(THF)x
8
16 h, -NaBr
1
t
/
(19)
2[(THF)BrNi]2(µ-SSi Bu3)2
8
Rapid desolvation prevented standard characterization of the
compound, but a DCl/D2O degradation in D3COD afforded
a 1.0:1.1 ratio of Bu3SiS/THF, hence the formulation.
Dissolution of 8 in benzene-d6 generated a purple solution
of 7, and a white precipitate, probably NiBr2, was noted.
Ligand disproportionation was occurring, and once noted in
polar solvents, it was observed in the formation of 5-NiBr
t
Inorganic Chemistry, Vol. 45, No. 2, 2006 615

Sydora et al.
Table 4. Comparative Interatomic Distances (Å) and Angles (deg) for [Fe(µ-Cl)(µ-SSitBu3)]12(C6H6)7 (5-FeCl), [Co(µ-Cl)(µ-SSitBu3)]12(C6H6)6
(5-CoCl), [Fe(µ-Br)(µ-SSitBu3)]12 (5-FeBr), [Ni(µ-Br)(µ-SSitBu3)]12 (5-NiBr), and [Fe(µ-I)(µ-SSitBu3)]14 (6-FeI)
5-FeCl
5-CoCl
5-FeBr
5-NiBr
6-FeI
M‚‚‚Mave
X‚‚‚X
3.127(44)
9.484(29) (ave)
3.115(40)
9.614(27) (ave)
3.171(51)
9.483(22) (ave)
3.314(41)
10.750(8)
9.583(8)
3.238(46)a
13.469(2)
12.402(2)
11.499(2)
11.101(2)
2.672(11)
2.342(7)
M-Xave
M-Save
2.338(9)
2.340(9)
2.310(11)
2.305(9)
2.474(9)
2.339(9)
2.440(9)
2.276(53)
M-X-Mave
M-S-Mave
X-M-Xave
S-M-Save
83.9(15)
83.9(14)
103.5(13)
124.8(6)
93.3(5)
84.8(13)
85.0(12)
107.8(13)
126.6(15)
91.9(4)
79.7(17)
85.4(18)
103.6(15)
127.4(8)
94.7(7)
85.5(15)
93.9(11)
103.3(1)
137.0(26)
88.8(17)
118.4(21)
74.6(13)
87.4(16)
101.6(7)
126.8(36)b
96.9(8)
c
X-M-Save
d
X-M-Save
121.0(14)
119.7(12)
117.7(13)
116.2(17)e
a Individual d(Fe‚‚‚Fe): 3.183(4), 3.217(4), 3.270(4), 3.280(4) Å. b The S(3)-Fe(4)-S(4) angle nearest the foci of the ellipse is 132.13(17)°, whereas the
remaining S-Fe-S angles are 124.6(2)°, 126.05(16)°, and 124.49(16)°. c Endo X-M-S angles. d Exo X-M-S angles. e The S(3)-Fe(4)-I(4) and S(4)-
Fe(4)-I(3) angles nearest the foci of the ellipse are 112.85(12)° and 115.35(16)°, respectively, whereas the remaining S-Fe-I angles average 117.1(7)°.
Figure 5. Molecular view of pseudo-D6d [Fe(µ-Cl)(µ-SSitBu3)]12(C6H6)7
Figure 6. Molecular view of pseudo-D6d [Fe(µ-Br)(µ-SSitBu3)]12 (5-
(5-FeCl).
FeBr).
could not be modeled. The final model reflects only the
whose metric parameters are very close to 5-CoCl, which
ordered part of the molecule devoid of solvent contribution.
Although 5-FeCl, 5-CoCl, and 5-FeBr crystallize in a
primitive tetragonal cell (the asymmetric unit is a quarter
wheel), their arrangement within the cell is essentially body-
centered, with two wheels per unit cell. Figure 9 shows the
columnar stacking of the 5-FeBr wheel, which includes a
channel down the center of the wheels and a smaller one
between stacks consisting of the alternating square arrays
of wheels. Benzenes have been located in the large channel
and feature edge-to-face stacking,20,21 as previously shown.7
The lattice parameters of 5-FeCl and 5-CoCl are virtually
identical (within ∼0.5%), reflecting the similarity in radii
between Fe (rcov(Fe) ) 1.17, rcov(Co) ) 1.16 Å) and Co.
The switch from Cl to Br causes a very modest increase in
has been illustrated previously.7 The peripheral carbons in
t
the Bu groups of pseudo-D6d 5-FeBr were refined isotro-
pically, as Figure 6 shows, and yet no disorder model could
be found, which is probably why the GOF is so poor. In
addition, the unit cell had two cavities of disordered benzenes
of solvation that could not be modeled; hence, refinement
required the use of SQUEEZE,19 and the final model only
reflects the ordered part of the cell. The poor data quality of
pseudo-C6V 5-NiBr (Figure 7) necessitated the use of
geometric constraints in the tBu3SiS groups, and SQUEEZE
was also needed, since the two cavities in the cell possessed
large amounts of disordered solvent molecules that could
not be modeled. The ellipse, 6-FeI, possesses idealized C2h
symmetry, as Figure 8 reveals, and also required isotropic
refinement of the tBu groups and employment of SQUEEZE
because disordered solvent molecules in the cell cavities
(20) Mecozzi, S.; West, A. P.; Dougherty, D. A. Proc. Natl. Acad. Sci.
U.S.A. 1996, 93, 10566-10571.
(21) (a) Klebe, G.; Diederich, F. Philos. Trans. R. Soc. London A 1993,
345, 37-48. (b) Bacon, G. E.; Curry, N. A.; Wilson, S. A. Proc. R.
Soc. London A 1964, 279, 98-110.
(19) Vandersluis, P.; Spek, A. L. Acta Crystallogr. 1990, A46, 194-201.
616 Inorganic Chemistry, Vol. 45, No. 2, 2006

Aggregation of [(tBu3SiS)MX]
slightly shorter and the “hinges” of the tetrahedral swing the
tBu3SiS groups very slightly more “up and down” relative
to the M12 plane, thereby increasing c at the expense of a )
b.
The tetrahedra must “flex” in response to the d(M-X)
because the transannular average d(X‚‚‚X) of 5-FeCl and
5-FeBr are essentially identical despite major differences
in covalent radii (rcov(Cl) ) 0.99, rcov(Br) ) 1.14 Å). The
greater average d(Fe-Br) of 2.474(9) Å vs the average
d(Fe-Cl) of 2.338(9) Å translates to a smaller average
X-Fe-X angle and a larger average S-Fe-S angle for
5-FeBr. The largest average X-M-X angles belong to
5-CoCl, which has the shortest average d(M-X), as
expected. In general, the wheels are remarkably similar, and
the changes in distances and angles between them are quite
subtle, as Table 3 reveals. Although the originally desired
helical polymers of “tBu3SiSMX” have not formed, the “up
and down” displacement of the tBu3SiS groups is essentially
what was expected. What was not expected was the
substantial degree to which the tetrahedral centers would
angularly distort in order to form these cyclic oligomers while
maintaining normal d(M-X) and d(M-S).
Figure 7. Molecular view of pseudo-C6V [Ni(µ-Br)(µ-SSitBu3)]12
(5-NiBr).
The nickel wheel, 5-NiBr, is somewhat of an oddity
because its secondary structure is subtly different than its
congeners. As Figure 7 depicts, 5-NiBr possesses idealized
C6V symmetry, with six alternating bromides pointing
more toward the center of the wheel than the other six
(d(Br‚‚‚Br) ) 9.583(8) vs 10.750(8) Å), which are more
vertical with respect to the wheel plane. If there were
questions regarding the templating of wheel formation with
benzene, the formation of 5-NiBr from hexane solution
would appear to dismiss that view. The wheel is still 12-
membered, but now the Ni center cannot accommodate the
bigger halide with a small Br-Ni-Br angle (cf 5-FeBr),
whose average is 85.5(15)°, the largest of all the cyclic
oligomers. The result is desymmetrization, a longer
d(M‚‚‚M) of 3.314(41) Å, and a change in packing that may
derive from the “lopsided” wheel. There are still two
molecules per unit cell, but 5-NiBr crystallizes in a primitive
hexagonal space group (P63mc) whose cell is roughly 10%
larger than the tetragonal ones. The ABAB ordering of the
hexagonal cell might be expected to pack more efficiently
than the pseudo-body-centered arrangements of 5-CoCl,
5-FeCl, and 5-FeBr, but the above factorssprincipally the
larger wheel size and accompanying solvent packingsmust
counteract a standard packing argument. The hexagonal
packing generates a different columnar array than the
tetragonal cells. One small, smooth channel is comprised of
the stacked trigonal holes that remain after ABAB packing,
and another channel is composed of alternating wheel centers
and trigonal holes. While templating wheel formation with
benzene can be discounted on the basis of the existence of
5-NiBr, the solvent change may be responsible for the shift
from a pseudo-body-centered to hexagonal packed arrange-
ment.
Figure 8. Molecular view of pseudo-C2h [Fe(µ-I)(µ-SSitBu3)]14 (6-FeI).
the wheel size, which is barely evident in the average
Fe‚‚‚Fe distance: 5-FeBr, (3.171(51) Å) > 5-FeCl
(3.127(44) Å) > 5-CoCl (3.115(40) Å). Curiously, while
the a ) b unit cell parameters increase from 5-CoCl
(23.287(3) Å) to 5-FeCl (23.3716(7) Å) to 5-FeBr
(24.542(5) Å) in line with the increasing sum of the covalent
radii, the c parameter manifests the opposite trend. Although
it is difficult to assay the role solvent plays in the unit cell
parameters, it is perhaps the connectivity of the wheels that
dictates this interplay. The wheels may be viewed as being
composed of (S‚‚‚X) edge-shared distorted tetrahedra, whose
inner X12 “ring” features six halides above and below the
M12 plane, hence the D6d idealized symmetry. As the
d(M-X) distances get shorter, the M‚‚‚M distances get very
A diversion from the 12-membered rings occurs for M )
Fe and X ) I in the form of the ellipse, 6-FeI shown in
Figure 8. The accompanying change from tetragonal to
Inorganic Chemistry, Vol. 45, No. 2, 2006 617

Sydora et al.
Figure 9. Columnar stacking of [Fe(µ-Br)(µ-SSitBu3)]12 (5-FeBr), with a dashed line for the top of the unit cell.
Figure 10. Columnar stacking of [Fe(µ-I)(µ-SSitBu3)]14 (6-FeI), with a dashed line for the top of the unit cell showing the a and c lengths.
orthorhombic cell doubles the number of molecules in the
cell to four but leaves intact the pseudo-body-centered
arrangement. Figure 10 illustrates the base-centered cell, but
by rotating the top of the cell by roughly 45°, the packing
can be seen as essentially the same as in 5-FeCl, 5-FeBr,
and 5-CoCl, and the same two types of channelssdown
and between the wheelssare evident. The ring has been
opened up to accommodate the d(FeI)ave ) 2.672(11) Å that
618 Inorganic Chemistry, Vol. 45, No. 2, 2006

Aggregation of [(tBu3SiS)MX]
Figure 11. Plot of øMT vs T for [Fe(µ-Cl)(µ-SSitBu3)]12 (5-FeCl) from
5 to 300K (1 T applied field).
Figure 13. Plot of ø′MT vs T for [Fe(µ-I)(µ-SSitBu3)]14 (6-FeI) from 5
to 300K, where ø′M is the molar in-phase ac susceptibility measured at
1000 Hz.
Figure 12. Plot of øMT vs T for [Fe(µ-Br)(µ-SSitBu3)]12 (5-FeBr) from
5 to 300K (1 T applied field).
Figure 14. Plot of ø′MT vs T for [Co(µ-Cl)(µ-SSitBu3)]14 (5-CoCl) from
5 to 300K, where ø′M is the molar in-phase ac susceptibility measured at
1000 Hz.
are substantially longer than d(FeS)ave ) 2.342(7) Å, but
the trans annular d(I‚‚‚I) are still not much different:
11.101(2), 11.499(2), 12.402(2), and 13.469(2) Å. As Table
4 details, the tetrahedra nearest the foci exhibit effects due
to additional strain, as the wheel kinks to adopt its secondary
structure.
and continues until the lowest temperature of 5 K is reached.
Complexes 6-FeI and 5-CoCl (Figures 13 and 14) exhibit
a similar response in their ø′MT vs T plots (ø′M is the molar
in-phase ac susceptibility). At 300 K, 6-FeI and 5-CoCl
have ø′MT values of 23 (µeff ≈ 15 µB) and 29 cm3‚K‚mol-1
(µeff ≈ 15.5 µB), respectively. These values remain essentially
unchanged with decreasing temperature until below 20 K
where a sharp decrease is noted.
A quantitative analysis of the intermolecular magnetic
exchange interactions between either the Fe(II) or Co(II) ions
in these four complexes is not possible due to the fact that
magnetic exchange interactions between pairs of Co(II) or
Fe(II) ions are anisotropic. Tetrahedrally coordinated Fe(II)
and Co(II) ions generally have large g-values where there is
considerable g-tensor anisotropy. The presence of appreciable
orbital angular momentum at these ions leads to anisotropic
magnetic exchange interactions between metal centers. The
simplifying models which describe the exchange as being
in either the isotropic, Ising, or XY limits are inappropriate.
6. Magnetism of the Cyclic Oligomers. Variable-tem-
perature magnetic susceptibility data were collected for
microcrystalline samples of 5-FeCl, 5-FeBr, 5-CoCl, and
6-FeI. Fear of persistent contaminants in the nickelous
bromide wheel (5-NiBr) obviated characterization of its
magnetism. Plots of øMT vs T for the cyclic oligomers are
depicted in Figures11-14. Upon decreasing the temperature
of 5-FeCl (Figure 11), the product øMT remains essentially
unchanged at a value of 41 cm3‚K‚mol-1 (µeff ≈ 18 µB) from
300 to 50 K, whereupon øMT slightly increases to a value of
47 cm3‚K‚mol-1 (µeff ≈ 19.5 µB). Below 15 K, the value of
øMT sharply decreases. The value of øMT for complex
5-FeBr (Figure 12) also remains essentially unchanged from
a value of 28 cm3‚K‚mol-1 (µeff ≈ 15.5 µB) at 300 K down
to 100 K, whereupon a smooth decrease of øMT is observed
Inorganic Chemistry, Vol. 45, No. 2, 2006 619

Sydora et al.
Detailed information on the single-ion anisotropy of each
unique metal center involved in the magnetic exchange would
be needed before any specific analysis of susceptibility data
for these complexes could be attempted. The situation is
further complicated by the existence of several unique Fe
or Co ions in the asymmetric unit of each crystal structure.
There are three unique Fe(II) or Co(II) atoms in complexes
5-FeCl, 5-FeBr, and 5-CoCl, and 6-FeI has four unique
Fe(II) atoms, each requiring its own g- and J-tensor. In
principle, one could extract single-ion anisotropy parameters
by doping a diamagnetic host, i.e., a wheel composed of
[ZnCl]12, with a small amount of Fe(II) or Co(II), but such
hosts have not been prepared despite considerable effort.
Single-crystal high-frequency EPR and magnetometry ex-
periments could then be performed, and a similar analysis
has recently been performed on 1-D Co(II) polymers.22
Attempts to fit øM vs T plots with the Heisenberg infinite
chain model23 yielded consistent results, despite the intrinsic
simplicity of the model. Reasonable fits of 5-FeBr and
5-CoCl were generated, and these wheels were weakly
antiferromagnetic.6 In modest contrast, the fit of 5-FeCl was
closest to that of a simple paramagnet. Although the
Heisenberg chain model is strictly inappropriate, the interac-
tions between crystallographically distinct metal centers in
5-FeBr, 5-CoCl, and 5-FeBr must be very similar in order
for it to have provided a reasonable approximation of each
wheel’s magnetism. Note that certain Fe(II) centers in 6-FeI
are significantly different, and here the chain model gave
no reasonable fit.
Figure 15. Plot of the molar in-phase (ø′M) and molar out-of-phase (ø′′M)
components for the ac susceptibility vs T for [Fe(µ-Cl)(µ-SSitBu3)]12 (5-
FeCl). The measurement was performed with a 1 G field oscillating at the
frequencies 1000 (squares), 500 (circles), and 50 Hz (triangles). The solid
lines are meant to quide the eyes and do not represent a theoretical fit.
Out-of-phase ac-susceptibility data were collected for all
four complexes, and the results are depicted in Figures15-
17. The iron chloride (5-FeCl) and iodide (6-FeI) com-
plexes both show out-of-phase ac responses beginning at ca.
5 K that are not frequency dependent. The values of ø′′M
are essentially superimposable upon one another, despite the
different frequencies probed (1000, 500, and 50 Hz). The
cobalt chloride species (5-CoCl) exhibits a similar response;
however, the temperature at which the ø′′M signal first
appeared is ca. 20 K. The iron bromide wheel (5-FeBr) did
not exhibit an out-of-phase component (not shown). The
appearance of an out-of-phase peak in a plot of ø′′M vs T is
often associated with single-molecule magnetism,26,27 but it
is not likely that this is the case here. The onset of the non-
frequency-dependent ø′′M peak is accompanied by a cusp in
the corresponding ø′M vs T plots. Both observations suggest
that there is likely a phase change at the temperatures where
the out-of-phase feature is observed. Specific heat measure-
ments would need to be performed to verify and ascertain
the nature of this phase change.
Despite quantitative difficulties, one can make the qualita-
tive observation that magnetic exchange interactions over
the temperature range examined appear to be weak since
there are only small variations of the product øMT until low
temperatures. With the exception of 5-FeCl, all interactions
appear to be very weakly antiferromagnetic. Below 15 K,
there is a precipitous drop of øMT toward zero. It is likely
that this drop in the øMT value at low temperatures is due to
zero-field splitting effects since tetrahedral Fe(II) and
Co(II) ions generally exhibit appreciable single-ion zero-field
interactions. It is intreresting that 5-FeCl shows a weak
ferromagnetic interaction between Fe(II) ions since it was
anticipated that the insulating properties of the ligands would
minimize intermolecular magnetic exchange interactions. It
may be possible that spin canting is the source of this weak
ferromagnetic interaction.24 This phenomenon has been
observed in a number of one-dimensional transition metal
polymers.25
(22) (a) Caneschi, A.; Gatteschi, D.; Lalioti, N.; Sangregorio, C.; Sessoli,
R.; Venturi, G.; Vindigni, A.; Rettori, A.; Pini, M. G.; Novak, M. A.
Angew. Chem., Int. Ed. 2001, 40, 1760-1762. (b) Caneschi, A.;
Gatteschi, D.; Lalioti, N.; Sorace, L.; Tangoulis, V.; Vindigni, A.
Chem. Eur. J. 2002, 8, 286-292.
(23) (a) Smith, T.; Friedberg, S. A. Phys. ReV. 1968, 176, 660-665. (b)
Dingle, R.; Lines, M. E.; Holt, S. L. Phys. ReV. 1969, 187, 643-648.
(c) Wagner, G. R.; Friedberg, S. A. Phys. Lett. 1964, 9, 11-13.
(24) Carlin, R. L. Magnetochemistry; Springer-Verlag: Berlin; New York,
1986.
(25) (a) Herweijer, A.; de Jonge, W. J. M.; Botterman, A. C.; Bongaarts,
A. L. M.; Cowen, J. A. Phys. ReV. B. 1972, B5, 4618. (b) Kopinga,
K.; van Vlimmeren, Q. A. G.; Bongaarts, A. L. M.; de Jonge W. J.
M. Physica 1977, B86-88, 671.
Discussion
t
1. Synthetic Investigations. 1.1. Bu3SiSH. Two inde-
t
t
pendent syntheses of Bu3SiSH and Bu3SiSNa have been
(26) Molecular Magnetism; Kahn, O., Ed.; VCH: New York, 1993.
(27) (a) Christou, G.; Gatteschi, D.; Hendrickson, D. N.; Sessoli, R. MRS
Bull. 2000, 25, 66-71. (b) Gatteschi, D.; Sessoli, R. J. Magn. Magn.
Mater. 2004, 272, 1030-1036. (c) Gatteschi, D.; Caneschi, A.; Pardi,
L.; Sessoli, R. Science 1994, 265, 1054-1058.
620 Inorganic Chemistry, Vol. 45, No. 2, 2006

Aggregation of [(tBu3SiS)MX]
Figure 16. Plot of the molar in-phase (ø′M) and molar out-of-phase (ø′′M)
components for the ac susceptibility vs T for [Fe(µ-I)(µ-SSitBu3)]14 (6-
FeI). The measurement was performed with a 1 G field oscillating at the
frequencies 1000 (squares), 500 (circles), and 50 Hz (triangles). The solid
lines are meant to quide the eyes and do not represent a theoretical fit.
Figure 17. Plot of the molar in-phase (ø′M) and molar out-of-phase (ø′′M)
components for the ac susceptibility vs T for [Co(µ-Cl)(µ-SSitBu3)]14
(5-CoCl). The measurement was performed with a 1 G field oscillating at
the frequencies 1000 (squares), 500 (circles), and 50 Hz (triangles). The
solid lines are meant to quide the eyes and do not represent a theoretical
fit.
detailed in accordance with brief reports by Wiberg.8
Utilization of the triflate, tBu3SiOTf,11 and anhydrous NaSH
toward thiol production is highly recommended. While
timewise the reaction sequences are similar, the simplicity
and high yields achieved via the triflate render the disulfide
route uncompetitive. Of course, the intermediate tBu3SiSSSit-
Bu3 may itself be useful as a reagent, but it can be generated
easily upon oxidation of the thiolate.
1.2. Oligomer Precursors and [(tBu3SiS)M]2(µ-SSitBu3)2
(M ) Fe, 12; Ni, 7). Precursors to the wheel and ellipse are
generally very simple coordination compounds that hold few
surprises. All of the metal centers are tetrahedral and linked
via thiolate bridges, and while the oligomeric nature of 3-Br
and, presumably 3-Cl, are modestly interesting, the structure
of 3-Br reveals no particular geometric feature of signifi-
cance.
The most interesting small aggregates discovered during
this investigation are the dimers, [(tBu3SiS)M]2(µ-SSitBu3)2
(M ) Fe, 12; Ni, 7). Possessing pseudo-trigonal M(II) centers,
these compounds are uncommon, and the modest pyramidal
distortion of each Fe in 12, while probably steric in origin,
renders its structure unique. Likewise, a literature search has
not revealed other Ni(II) thiolate dimers of a pseudo-trigonal
nature. Although Power and co-workers synthesized a
number of three-coordinate metal dimers [(ArS)M]2(µ-SAr)2
(M ) Mn, Fe, Co; Ar ) 2,4,6-tBu3C6H2),14 no comparable
work has been found for nickel. The reactivity and properties
of these dimers will be investigated further.
1.3. Wheels and Ellipse. Lowering the coordination
number through desolvationswhether in situ or via well-
characterized precursorssis a reasonable approach to oli-
gomer formation when the metal-solvent bonds are weak,
as they are in these first row cases that involve THF. Modest
thermolysis conditionsstypically <120 °Cswere quite ef-
fective in removal of THF and any remaining salt that was
bound with the aid of solvent. It is unknown what the
mechanism of aggregation is upon desolvation and dissolu-
tion in benzene (or hexane, as in the case of 5-NiBr), but
it is convenient to think of the oligomerization of “tBu3-
SiSMX”. As the 2-coordinate species dimerizes, then adds
another segment, etc., the growing chain has two possibilities
for growth: (1) one in which a helical chain is formed as
the bulky µ-SSitBu3 unit moves from one bridging position
to the next by ∼90° jumps until all four positions (looking
down a chain) are used and the sequence is repeated
indefinitely and (2) one in which the bulky µ-SSitBu3 unit
alternates between two positions, causing the growing chain
to bend due to growing peripheral steric influences, and
ultimately generating the ring. If one accepts the premise
that the wheels are unlikely to be entropically favored with
respect to a distribution of oligomers, then perhaps greater
relief of tBu group steric interactions in the wheels and ellipse
are at the origin of ring formation. Unfortunately, with a
limited ability to characterize solution phase aggregates, the
Inorganic Chemistry, Vol. 45, No. 2, 2006 621

Sydora et al.
mechanism of ring formation, and even the thermodynamics
of the system, must remain the subject of speculation.
early thiolate-based rings were comprised of square planar
metal centers, such as the hexamer [Ni(µ-SEt)2]6.48 Related
tetrameric Ni(II) thiolates,49 pentamers50,51 and octamers52s
including those of Cu(II)53,54sare also known. It is clear that
the steric bulk of Bu3SiS must play a significant role in
allowing “(tBu3SiS)2Ni” to diimerize but not aggregate
further.
In the past 15 years, there have been numerous reports of
ring or wheel compounds, yet those described herein are quite
unusual in that a cyclic array of tetrahedra, bent at the M(µ-
X)(µ-SSitBu3) “hinges”, comprise the molecules. Most ver-
sions consist of linked octahedra containing M(III) (M )
Cr,28-31 Fe)32-38 or M(II) (M ) Mn,39,40 Co,41-43 Ni)44,45 ions
coordinated by “hard” oxygen- or nitrogen-based ligands,
including many that are bidentate or polydentate. Excluding
polyoxoanions, few rings exist in group 546 or in the second
(e.g., â-MoCl4 or [MoCl2(µ-Cl)2]6)47 or third rows. Some
t
In terms of rings composed of tetrahedra, there are a few
for comparison. [Fe6S6I6]2- is a mixed-valence iron cluster,55
whose tetrahedra are linked via adjacent edges that utilize
µ3-S bridges connecting FeI units in a D3d barrel arrangement.
A spectacular ring structure based on linked tetrahedra is
the mixed-valence iron-sulfur cluster [Na2Fe18S30],8- in
which 14 FeS4 units share tetrahedral edges to form a toroid
and 4 additional iron centers bind the ring to form incomplete
cubes.56 The most relevant predecessor is the dodecanuclear
wheel, [Fe(µ-SePh)2]12,57 which is the first example of the
type of wheel described herein, having µ-SePh ligands that
alternate above and below the ring plane.
(28) McInnes, E. J. L.; Anson, C.; Powell, A. K.; Thomson, A. J.;
Poussereau, S.; Sessoli, R. J. Chem. Soc., Chem. Commun. 2001, 89-
90.
(29) (a) Atkinson, I. M.; Benelli, C.; Murrie, M.; Parsons, S.; Winpenny,
R. E. P. J. Chem. Soc., Chem. Commun. 1999, 285-286. (b) Larsen,
F. K.; Overgaard, J.; Parsons, S.; Rentschler, E.; Smith, A. A.; Timco,
G. A.; Winpenny, R. E. P. Angew. Chem., Int. Ed. 2003, 42, 5978-
5981. (c) Larsen, F. K.; McInnes, E. J. L.; El Mkami, H.; Rajaraman,
G.; Rentschler, E.; Smith, A. A.; Smith, G. M.; Boote, V.; Jennings,
M.; Timco, G. A.; Winpenny, R. E. P. Angew. Chem., Int. Ed. 2003,
42101-105.
(30) Ge´rbe´le´u, N. V.; Struchkov, Y. T.; Timco, G. A.; Batsanov, A. S.;
Indrichan, K. M.; Popovich, G. A. Dokl. Akad. Nauk. SSSR 1990, 313,
1459-1462.
(31) Eshel, M.; Bino, A.; Felner, I.; Johnston, D. C.; Luban, M.; Miller, L.
L. Inorg. Chem. 2000, 39, 1376-1380.
(32) (a) Taft, K. L.; Lippard, S. J. J. Am. Chem. Soc. 1990, 112, 9629-
9630. (b) Taft, K. L.; Delfs, C. D.; Papaefthymiou, G. C.; Foner, S.;
Gatteschi, D.; Lippard, S. J. J. Am. Chem. Soc. 1994, 116, 823-831.
(33) Watton, S. P.; Fuhrmann, P.; Pence, L. E.; Caneschi, A.; Cornia, A.;
Abbati, G. L.; Lippard, S. J. Angew. Chem., Int. Ed. Engl. 1997, 36,
2774-2776.
(34) Jones, L. F.; Batsanov, A.; Brechin, E. K.; Collison, D. Helliwell,
M.; Mallah, T.; McInnes, E. J. L.; Piligkos, S. Angew. Chem., Int.
Ed. 2002, 41, 4318-4321.
(35) (a) Benelli, C.; Parsons, S.; Solan, G. A.; Winpenny, R. E. P. Angew.
Chem., Int. Ed. Engl. 1996, 35, 1825-1828. (b) Abu-Nawwas, A. A.
H.; Cano, J.; Christian, P.; Mallah, T.; Rajaraman, G.; Teat, S. J.;
Winpenny, R. E. P.; Yukawa, Y. Chem. Commun. 2004, 314-315.
(36) Raptopoulou, C. P.; Tangoulis, V. Devlin, E. Angew. Chem., Int. Ed.
2002, 41, 2386-2389.
While the wheels 5-MX (MX ) FeCl, FeBr, CoCl, NiBr)
fit as a subset within the structural types mentioned above,
the ellipse 6-FeI is quite unusual, and related cyclic arrays
have not been found in the literature. In fact, very few ring
structures possessing greater than 12 metal centers have been
noted,33,34,40,45,56 and while [Na2Fe18S30]8- is elliptical in
shape, µ-sulfide bonds to counterions render it quite distinct
from neutral 6-FeI.
2. Wheel Magnetic Studies. For three cases, 5-FeBr,
6-FeI, and 5-CoCl, the wheels exhibited weak antiferro-
magnetic behavior consistent with poorly coupled tetrahedral
metal centers. Despite the apparent high symmetry of the
rings, their intrinsic complexity precludes a standard analysis.
The ferrous chloride 5-FeCl manifested modest ferromag-
netic coupling, and the nature of the system suggests that a
spin-canting mechanism may be operational24 since the steric
shielding of the system would tend to minimalize any
intermolecular interactions. Susceptibility studies that were
conducted in ac mode revealed features of 5-FeCl, 6-FeI,
and 5-CoCl that pointed toward molecular magnetism.26,27
However, the modest intensity and apparent lack of a
(37) (a) Caneschi, A.; Cornia, A.; Fabretti, A. C.; Gatteschi, D. Angew.
Chem. Int. Ed. 1999, 38, 1295-1297. (b) Abbati, G. L.; Caneschi,
A.; Cornia, A.; Fabretti, A. C.; Gatteschi, D. Inorg. Chim. Acta 2000,
297, 291-300. (c) Caneschi, A.; Cornia, A.; Lippard, S. J. Angew.
Chem., Int. Ed. Engl. 1995, 34, 467-469. (d) Caneschi, A.; Cornia,
A.; Fabretti, A. C.; Foner, S.; Gatteschi, D.; Grandi, R.; Schenetti, L.
Chem. Eur. J. 1996, 2, 1379-1387.
(38) Saalfrank, R. W.; Bernt, I.; Uller, E.; Hampel, F. Angew. Chem., Int.
Ed. Engl. 1997, 36, 2482-2485.
(39) (a) Abbati, G. L.; Cornia, A.; Fabretti, A. C.; Caneschi, A.; Gatteschi,
D. Inorg. Chem. 1998, 37, 1430-1431. (b) Caneschi, A.; Gatteschi,
D.; Laugier, J.; Rey, P.; Sessoli, R.; Zanchini, C. J. Am. Chem. Soc.
1988, 110, 2795-2799.
(48) (a) Woodward, P.; Dahl, L. F.; Abel, E. W.; Crosse, B. C. J. Am.
Chem. Soc. 1965, 87, 5251-5253. (b) For a Pd analogue, see:
Kunchur, N. R. Acta Crystallogr. Sect. B 1968, B24, 1623-1634.
(49) Gaete, W.; Ros, J.; Solans, X.; Font-Altaba, M.; Brianso´, J. L. Inorg.
Chem. 1984, 23, 39-43.
(40) Liu, S.X.; Lin, S.; Lin, B.-Z.; Lin, C.-C.; Huang, J-Q. Angew. Chem.,
Int. Ed. 2001, 40, 1084-1087.
(50) Koo, B.-K.; Block, E.; Kang, H.; Liu, S.; Zubieta, J. Polyhedron 1988,
7, 1397-1399.
(41) Beattie, J. K.; Hambley, T. W.; Klepetko, J. A.; Masters, A. F.; Turner,
P. J. Chem. Soc., Chem. Commun. 1998, 45-46, 14.
(42) Brechin, E. K.; Cador, O.; Caneschi, A.; Cadiou, C.; Harris, S. G.;
Parsons, S.; Vonci, M.; Winpenny, R. E. P. J. Chem. Soc., Chem.
Commun. 2002, 1860-1861.
(43) Jones, P. L.; Byrom, K. J.; Jeffery, J. C.; McCleverty, J. A.; Ward,
M. D. J. Chem. Soc., Chem. Commun. 1997, 1361-1362.
(44) (a) Blake, J.; Grant, C. M.; Parsons, S.; Rawson, J. M.; Winpenny, R.
E. P. Chem. Commun. 1994, 2363-2364. (b) Andres, H.; Basler, R.;
Blake, A. J.; Cadiou, C.; Chaboussant, G.; Grant, C. M.; Gudel, H.
U.; Murrie, M.; Parsons, S.; Paulson, C.; Semadini, F.; Villar, V.;
Wernsdorfer, W.; Winpenny, R. E. P. Chem. Eur. J. 2002, 8, 4867-
4876.
(51) Kriege, M.; Henkel, G. Z. Naturforsch. B. 1987, 42, 1121-1128.
(52) Dance, I. G.; Scudder, M. L.; Secomb, R. Inorg. Chem. 1985, 24,
1201-1208.
(53) Galy, J.; Mosset, A.; Grenthe, I.; Puigdome`nech, I.; Sjo¨berg, B.; Hulte´n,
F. J. Am. Chem. Soc. 1987, 109, 380-386.
(54) a) Ardizzoia, G. A.; Angaroni, M. A.; La Monica, G.; Cariati, F.;
Moret, M.; Masciocchi, N. J. Chem. Soc., Chem. Commun. 1990,
1021-1023. (b) Ardizzoia, G. A.; Angaroni, M. A.; La Monica, G.;
Cariati, F.; Cenini, S.; Moret, M.; Masciocchi, N. Inorg. Chem. 1991,
30, 4347-4353.
(55) Saak, W.; Henkel, G.; Pohl, S. Angew. Chem., Int. Ed. Engl. 1984,
23, 150-151.
(45) Dearden, A. L.; Parsons, S.; Winpenny, R. E. P. Angew. Chem., Int.
Ed. 2001, 40, 151-154.
(46) Chen, Q.; Liu, S.; Zubieta, J. Inorg. Chem. 1989, 28, 4434-4436.
(47) Mu¨ller, U. Angew. Chem., Int. Ed. Engl. 1981, 20, 692-693.
(56) You, J.-F.; Synder, B. S.; Papaefthymiou, G. C.; Holm, R. H. J. Am.
Chem. Soc. 1990, 112, 1067-1076.
(57) Fenske, D.; Fischer, A. Angew. Chem., Int. Ed. Engl. 1995, 34, 307-
309.
622 Inorganic Chemistry, Vol. 45, No. 2, 2006

Aggregation of [(tBu3SiS)MX]
frequency dependence in the out-of-phase measurements
have effectively ruled this out. At the low temperatures the
phenomena were observed, the compounds were likely
undergoing phase changes
The study and synthesis of antiferromagnetic rings has
been pointed out as a significant area of research for
nanomagnetism in a recent review.58 The quantum effects
associated with these antiferromagnetic rings are expected
to be important. Quantum coherence is achieved when a
material can tunnel (or “oscillate”) between two states
without dissipating or absorbing energy from its environment.
A necessary precondition for quantum computing59 with
electronic spin states is their coherent evolution under
controlled perturbations. The electronic spins in molecular
magnets interact with their environment, which includes
nuclear spins, lattice vibrations, and other molecular magnets
in the crystal. These tend to destroy the coherence and, for
example, reduce the expected quantum oscillations between
“up” and “down” spin states to an “incoherent” tunneling
transition between these states. The quantum tunneling in
an antiferromagnetic ring is predicted to occur coherently.60,61
If there are an odd number of antiferromagnetically coupled
spin carriers in a ring, the compound will have an uncom-
pensated spin ground state.62 For example, a compound with
Figure 18. Plot of the reduced magnetization (M/âN), where M is the
molar magnetization, N is Avogadro’s number, and â is the Bohr magneton,
vs H (in Tesla) for [Co(µ-Cl)(µ-SSitBu3)]14 (5-CoCl). The data were
collected at 5.0 K.
In summary, only modest couplingssboth antiferro-
magnetic and ferromagneticswere observed for the wheels
and ellipse, presumably due to rather long intermetallic
distances. No further investigations were conducted.
1
thirteen antiferromagnetically coupled S ) /2 spin carriers
Experimental Section
1
will have a spin ground state of S ) /2. If the remaining
1. General Considerations. All manipulations were performed
using either glovebox or high-vacuum-line techniques. Hydrocarbon
solvents containing 1-2 mL of added tetraglyme and ethereal
solvents were distilled under nitrogen from purple sodium ben-
zophenone ketyl and vacuum-transferred from the same prior to
use. C6D6 was dried over activated 4 Å molecular sieves, vacuum-
transferred, and stored under N2. The compounds tBu3SiNa(THF)2,10
tBu3SiOTf,11,63 FeCl2(THF)2,64 FeBr2(THF)2,65 FeI2(THF)2,65 and
uncompensated spin (S ) 1/2 in this case) is forced to change
orientations from a “spin-up” to a “spin-down” (ms ) -1/2
1
to ms ) /2) then all the other spin carriers in the ring must
also reverse their spin in order to maintain the antiferro-
magnetic exchange interaction. At low enough temperatures,
this is expected to occur by a tunneling mechanism and also
to occur coherently. One can also still observe quantum
effects in even-numbered antiferromagnetic rings. If there
is a low-lying excited spin-state not far in energy from the
S ) 0 ground state, one can populate this excited state by
applying a strong magnetic field. At high enough magnetic
field, the excited state will cross over and become lower in
energy than the ground state. At low enough temperatures,
this crossover can occur via quantum tunneling. Gatteschi
and Lippard et al. have examined this for an Fe10 “ferric
wheel”.32 At certain intervals of applied magnetic field,
crossovers were observed from the ground state to low-lying
excited states. The cobalt wheel 5-CoCl was examined in
a similar way (Figure 18), and an essentially linear response
of M/âN vs H is observed. This is typical for an antiferro-
magnet, and no crossover was observed. It is likely that the
required magnetic field to induce a crossover is much higher
than the instrumentation utilized.
66
NiBr2(THF)2 were prepared according to literature procedures.
All glassware was oven-dried, and NMR tubes for sealed tube
experiments were additionally flame-dried under dynamic vacuum.
NMR spectra were obtained using INOVA-400 and Unity-500
spectrometers, and chemical shifts are reported relative to C6D6
(1H, 7.15; 13C{1H}, 128.39) and THF-d8 (1H, 1.73; 13C{1H}, 25.37).
Infrared spectra were recorded on a Nicolet Impact 410 spectro-
photometer interfaced to a Gateway PC. Elemental analyses were
performed by Oneida Research Services, Whitesboro, NY, or
Robertson Microlit Laboratories, Madison, NJ. Magnetic moments
were determined in C6D6 or THF-d8 at room temperature using
Evans’ method16 with an applied diamagnetic correction.
2. Procedures. 2.1. tBu3SiS-SSitBu3. A solution of S2Cl2 (1.24
mL, 15.50 mmol, ∼10 mL THF) was added dropwise to a solution
of tBu3SiNa(THF)2.6 (12.15 g, 29.64 mmol, ∼50 mL THF) cooled
to 0 °C. After addition was complete, the solution was allowed to
warm to 23 °C and stirred for 4.5 h. The solvent was removed in
vacuo, and the residue was extracted into pentane and filtered
through Celite. Solid product was obtained by cooling a concen-
(58) Gatteschi, D.; Sessoli, R. Angew. Chem., Int. Ed. 2003, 42, 268-297.
(59) (a) Leuenberger, M. N.; Loss, D. Nature 2001, 410, 789-793. (b)
DiVincenzo, D. P.; Loss, D. J. Magn. Magn. Mater. 1999, 200, 202-
218.
t
trated pentane solution to -78 °C to afford 5.06 g of Bu3SiS-
1
SSitBu3 (74% yield). H NMR (C6D6): δ 1.31 (s, CH3). 13C{1H}
(60) Meier, F.; Loss, D. Physica B: Condens. Matt. 2003, 329, 1140-
(63) Eaborn, C.; Saxena, A. K. J. Organomet. Chem. 1984, 271, 33-46.
(64) Herzog, S.; Gustav, K.; Kruger, E.; Oberender, H.; Schuster, R. Z.
Chem. 1963, 3, 428-429.
1141.
(61) Troiani, F.; Ghirri, A.; Affronte, M.; Carretta, S.; Santini, P.; Amoretti,
G.; Piligkos, S.; Timco, G.; Winpenny, R. E. P. Phys. ReV. Lett. 2005,
94, 207-208.
(62) Bartlett, B. M.; Nocera, D. G. J. Am. Chem. Soc. 2005, 127, 8985-
8993.
(65) Ittel, S. D.; English, A. D.; Tolman, C. A.; Jesson, J. P. Inorg. Chim.
Acta 1979, 33, 101-106.
(66) Casalnuovo, A. L.; RajanBabu, T. V.; Ayers, T. A.; Warren, T. H. J.
Am. Chem. Soc. 1994, 116, 9869-9882.
Inorganic Chemistry, Vol. 45, No. 2, 2006 623

Sydora et al.
1
NMR (C6D6): δ 26.36 (C(CH3)3), 31.18 (C(CH3)3). Anal. Calcd
cooled to -78 °C to yield 118 mg of 1-(THF)2 (46%). H NMR
for C12H27SiS: C, 62.2; H, 11.8. Found: C, 61.7; H, 11.3.
(C6D6): δ 2.84 (ν1/2 ≈ 40 Hz, CH3), 1.87 (ν1/2 ≈ 30 Hz, OCH2CH2),
t
t
4.68 (ν1/2 ≈ 80 Hz, OCH2CH2).
2.2. Bu3SiSH. 2.2.1. Substitution of Bu3SiOTf. A 25 mL
round-bottom flask attached to a reflux condenser was charged with
tBu3SiOTf (1.00 g, 2.869 mmol), anhydrous NaSH (0.161 g, 2.872
mmol), and THF (10 mL). The solution was refluxed under argon
for 3 d, and the solvent was removed in vacuo. The solid was then
extracted with pentane, and the solution was filtered, stripped of
2.6. (tBu3SiS)3Fe(THF) (1-(THF)2. A 25 mL flask was charged
with 0.250 g (0.692 mmol) of NaSSitBu3(THF)1.48, 37 mg of FeCl3
(0.23 mmol), and 15 mL of THF at -78 °C. The reaction mixture
was allowed to slowly warm to 23 °C over the course of 8 h and
was stirred for 1 h at 23 °C. The THF was removed, and the solid
was triturated with pentane (3 × 5 mL). Pentane (∼15 mL) was
added, and the mixture was filtered and washed. The solution was
filtered, the salt cake was washed (3 × 2 mL) with pentane, and
the extracts were reduced to ∼5 mL and cooled to -78 °C to
produce purple crystals of 2-THF (190 mg, 85%). 1H NMR (C6D6):
δ 21.4 (ν1/2 ≈ 1100 Hz). The compound was too thermally sensitive
to submit for analysis.
t
1
solvent, and sublimed to give Bu3SiSH (0.561 g, 84%). H NMR
(C6D6): δ -0.57 (s, 1H, SH), 1.10 (s, 27H, CH3). 13C{1H} NMR
(C6D6): δ 24.14 (C(CH3)3), 30.66 (C(CH3)3). IR (Nujol Mull, NaCl,
cm-1): 2178 (w), 1303 (w), 1192 (m), 1014 (m), 932 (m), 817 (s),
608 (m), 578 (m), 522 (m), 456 (m). Anal. Calcd for C13H27SiS:
C, 62.0; H, 11.7. Found: C, 62.2; H, 12.2.
2.2.2. LiAlH4 Reduction of tBu3SiS-SSitBu3. A heterogeneous
mixture of tBu3SiS-SSitBu3 (2.60 g, 5.615 mmol) and LiAlH4 (1.50
g, 39.52 mmol) in THF (60 mL) was refluxed under argon for 2 d.
The excess LiAlH4 was deactivated by H2O addition at 0 °C, and
the solution was acidified with aqueous 5 M HCl. The THF was
removed in vacuo and the product was extracted with Et2O and
dried over MgSO4. Solvent removal followed by sublimation gave
2.7. [Cl2Fe](µ-SSitBu3)2[FeCl(THF)]Na(THF)4 (3-Cl). A 50
mL flask was charged with NaSSitBu3(THF)1.48 (0.250 g, 0.692
mmol), FeCl2(THF)2 (0.188 g, 0.694 mmol), and THF (20 mL).
The pale yellow solution was stirred for 18 h, filtered, and the
solvent allowed to slowly evaporate to give crystalline 3-Cl (0.271
1
g, 74%). H NMR (THF-d8): δ 9.11 (υ1/2 ≈ 940 Hz). IR (Nujol
t
Mull, NaCl, cm-1): 1248 (w), 1192 (m), 1176 (m), 1018 (s), 932
(m), 915 (m), 868 (s), 817 (s), 724 (m), 679 (m), 613 (s), 570 (s),
515 (s), 460 (s), 423 (m). Anal. Calcd C28H62OSi2S2NaCl3Fe2
(desolvated): C, 43.3; H, 8.1; Cl, 13.7. Found: C, 42.5; H, 8.2; Cl,
13.4. µeff ) 4.8 µB at 295 K (Evans’ method in THF-d8). A quench
2.18 g of crystalline Bu3SiSH in 84% yield.
t
2.2.3. Protonation of Bu3SiSNa(THF)x. A 50 mL flask was
charged with tBu3SiSNa(THF)1.9 (3.043 g, 7.77 mmol) and attached
to a gas bulb. THF (20 mL) was added at -78 °C and the flask
was cooled to 77 K. HCl (538 Torr in 265.8 mL, 7.77 mmol) was
condensed into the flask, and the solution was slowly warmed to
23 °C. After the mixture was stirred for 30 min, the volatiles were
stripped and the solid was triturated three times with hexanes. The
crude solid was dissolved in hexane and filtered through a pad of
Celite. Volatiles were removed in vacuo to yield 1.52 g of crystalline
tBu3SiSH (84% yield).
1
with DCl/CD3OD afforded thiolate/THF ) 1.0:3.9 by H NMR
spectroscopic analysis.
2.8. [Br2Fe](µ-SSitBu3)2[FeBr(THF)]Na(THF)4 (3-Br). A 50
mL flask was charged with NaSSitBu3(THF)1.68 (0.325 g, 0.865
mmol), FeBr2(THF)2 (0.311 g, 0.864 mmol), and THF (20 mL).
The pale yellow solution was stirred for 18 h, filtered, and the
solvent allowed to slowly evaporate to give 0.446 g 3-Br (86%).
1H NMR (THF-d8): δ 12.46 (υ1/2 ≈ 280 Hz). IR (Nujol Mull, NaCl,
cm-1): 1260 (m), 1180 (m), 1057 (s), 1014 (s), 934 (m), 915 (m),
858 (s), 817 (s), 722 (w), 613 (s), 574 (s), 515 (s), 462 (s). Anal.
Calcd C32H70O2Si2S2Na Br3Fe2 (desolvated): C, 42.7; H, 7.9; Br,
21.3. Found: C 42.3, H 7.9, Br 21.8. µeff ) 4.5 µB at 295 K (Evans
method in THF-d8). A quench with DCl/CD3OD afforded thiolate/
2.3. tBu3SiSNa(THF)x (x ) 1.4-1.9). 2.3.1. Sodium Reduction
t
of Bu3SiSH. A 100 mL round-bottom flask was charged with
tBu3SiSH (2.40 g, 10.3 mmol), sodium metal (0.48 g, 20.9 mmol),
and THF (50 mL). The mixture was stirred overnight, filtered,
concentrated, crystallized at -78 °C, and dried in vacuo to afford
t
white, crystalline Bu3SiSNa(THF)1.40 (3.37 g, 92%). 1H NMR
(C6D6): δ 1.36 (s, 27 H, CH3), 1.41 (m, 4H, OCH2CH2), 3.61 (t,
4H, OCH2CH2). 13C{1H} NMR (C6D6): δ 25.83 (C(CH3)3), 32.07
(C(CH3)3). IR (Nujol Mull, NaCl, cm-1): 1180 (m), 1050 (m), 1013
(m), 932 (w), 912 (w), 817 (m), 603 (m), 588 (m), 529 (m), 461
(m). Anal. Calcd for C16H35ONaSiS (desolvated): C, 58.8; H, 10.8.
Found: C, 58.6; H, 10.9.
1
THF ) 1.0:3.9 by H NMR spectroscopic analysis.
2.9. cis-[(THF)IFe]2(µ-SSitBu3)2 (4). A 50 mL flask was charged
with NaSSitBu3(THF)1.68 (0.504 g, 1.34 mmol), FeI2(THF)2 (0.608
g, 1.34 mmol), and THF (25 mL). The yellow solution was stirred
for 24 h, filtered, and the solvent was removed. The yellow solid
was heated under vacuum at 98 °C for 2 h, extracted into benzene
(40 mL), and filtered. Slow evaporation of the solvent afforded 4
(0.485 g, 75%) after 5 days. 1H NMR (C6D6): δ 6.82 (υ1/2 ≈ 140
Hz). IR (Nujol Mull, NaCl, cm-1): 1243 (w), 1188 (m), 1176 (m),
1014 (s), 934 (s), 858 (s), 817 (s), 724 (w), 677 (m), 615 (s), 566
(s). Anal. Calcd C12H27SiSIFe (desolvated): C, 34.8; H, 6.6; I, 30.6.
Found: C, 33.4; H, 6.4; I, 27.7. µeff/Fe ) 4.8 µB at 295 K (Evans’
method in C6D6). A quench with DCl/CD3OD afforded thiolate/
2.3.2. Na/Hg Reduction of tBu3SiS-SSitBu3. A 100 mL round-
t
bottom flask was charged with Bu3SiS-SSitBu3 (1.42 g, 3.07
mmol), freshly made 0.9% Na/Hg amalgam (30.3 g, 11.9 mmol
Na°), and THF (60 mL). The heterogeneous mixture was stirred
for 3 days, filtered, and concentrated at -78 °C yielding tBu3SiSNa-
(THF)1.90 (1.32 g, 55%).
2.4. [(tBu3SiS)Fe]2(µ-SSitBu3)2 (12). A 50 mL flask was charged
t
with [{(Me3Si)2N}2Fe]2 (1.099 g, 2.918 mmol), Bu3SiSH (1.357
1
THF ) 1.0:0.9 by H NMR spectroscopic analysis.
g, 5.836 mmol), and 25 mL of benzene at 23 °C. Upon being stirred
for 30 min, the volatiles were removed and the brown solid was
subjected to dynamic vacuum for 3 h. The material was dissolved
in pentane, filtered, and cooled to -78 °C to provide 1.308 g of
orange 12 (86%). 1H NMR (C6D6, 23 °C): δ 1.91 (ν1/2 ) 200 Hz),
3.63 (ν1/2 ) 200 Hz). Anal. Calcd for C24H54Si2S2Fe: C, 55.56; H,
10.49. Found: C, 55.3; H, 10.6.
2.5. (tBu3SiS)2Fe(THF)2 (1-(THF)2. A 25 mL flask was charged
with 12 (200 mg, 0.193 mmol) and 15 mL of THF at -78 °C. The
solution was slowly brought to 23 °C, and the THF was removed.
Pentane (∼15 mL) was added, and the solution was filtered and
2.10. [Fe(µ-Cl)(µ-SSitBu3)]12(C6H6)7 (5-FeCl). A 50 mL flask
was charged with 3-Cl (0.271 g, 0.255 mmol) and heated under
vacuum at 80 °C for 1.5 h. The product was extracted with benzene
and filtered, and yellow crystalline 5-FeCl (0.082 g, 45%)
deposited after slow evaporation of the solvent (6 d). IR (Nujol
Mull, NaCl, cm-1): 1192 (m), 1178 (m), 1014 (s), 981 (m), 934
(s), 872 (w), 817 (s), 724 (w), 677 (s), 615 (s), 566 (s). Anal. Calcd
C12H27SiSClFe (desolvated): C, 44.6; H, 8.5; Cl, 11.0. Found: C,
43.9; H, 8.4; Cl, 9.9. A quench with DCl/CD3OD afforded thiolate/
1
C6H6 ) 1.0:0.6 by H NMR spectroscopic analysis.
624 Inorganic Chemistry, Vol. 45, No. 2, 2006

Aggregation of [(tBu3SiS)MX]
2.11. [Fe(µ-Br)(µ-SSitBu3)]12(C6H6)12 (5-FeBr). A 100 mL
flask was charged with NaSSitBu3(THF)1.40 (0.801 g, 2.25 mmol),
FeBr2(THF)2 (0.736 g, 2.28 mmol), and THF (50 mL). The mixture
was stirred at 23 °C for 24 h, and the solvent was removed to yield
a yellow solid. The solid was heated at 79 °C under vacuum for 2
h then extracted into benzene and filtered. After slow evaporation
of the solvent (5 days), yellow crystals of 5-FeBr formed (0.739
IR(Nujol Mull): 1189 (w), 1012 (m), 934 (m), 816 (m), 614 (s),
563 (s) cm-1. Anal. Calcd for C12H27SiSBrNi (desolvated): C, 38.9;
H, 7.4. Found: C, 33.5; H, 7.4.
3. Magnetic Susceptibility Measurements. Magnetic suscep-
tibility measurements were performed on Quantum Design mag-
netometers. Direct current (dc) susceptibilities were collected on a
model MPMS-5 SQUID magnetometer equipped with a 5.5 T
magnet. Alternating current (ac) susceptibilities were collected on
a model MPMS2 SQUID magnetometer. Both instruments operate
in the 2-400 K temperature range. All sample preparations and
manipulations were performed under an inert atmosphere to ensure
that the integrities of the air-sensitive samples were preserved. The
samples were either measured in a flame-sealed NMR tube or a
custom-machined, sealed Teflon capsule. Variable-temperature dc
susceptibility measurements were performed with an applied field
of 1 T. Variable-temperature ac susceptibility measurements were
performed with an oscillating ac field of 1 G and zero applied dc
field. The diamagnetic contribution from the sample container was
subtracted from the experimental data. Pascal’s constants24 were
also used to subtract the diamagnetic contributions, yielding
paramagnetic susceptibilities.
4. X-Ray Crystal Structure Determinations. 4.1. General. The
selected crystal (for 173 K data collection, it was immersed in
polyisobutylene) was placed in the goniometer head of a diffrac-
tometer equipped with a fine-focus molybdenum X-ray tube and
graphite monochromator. Preliminary diffraction data revealed the
crystal system (Tables 1 and 3), and a hemisphere routine was used
to collect the data. Precise lattice constants were determined from
a least-squares fit of 15 diffractometer-measured 2θ values. The
space group was determined, and after correction for Lorentz,
polarization, and background effects, unique data were judged
observed according to |Fo| > 2σ|Fo|. All heavy atoms were located
using direct methods, and all non-hydrogen atoms were revealed
by successive Fourier syntheses. Full matrix, least squares refine-
ments (minimization of Σw(Fo - Fc)2 where w is based on counting
statistics modified by an ignorance factor, w-1) with anisotropic
heavy atoms and all hydrogens included at calculated positions led
to the final model. Crystallographic data (CIF files) for certain
structures have been deposited with the Cambridge Crystallographic
Data Centre as supplementary publication nos. CCDC-203239
(5-FeCl), CCDC-203238 (5-FeBr), CCDC-203240 (6-FeI),
CCDC-220005 (5-CoCl), and CCDC-220006 (5-NiBr). Copies
of the data can be obtained free of charge on application to CCDC,
12 Union Road, Cambridge CB2 1EZ, UK (fax: (+44) 1223-
336-033; E-mail: deposit @ccdc.cam.ac.uk).
1
g, 72%). H NMR (C6D6): δ 4.84 (υ1/2 ≈ 60 Hz, tentative). IR
(Nujol Mull, NaCl, cm-1): 1190 (s), 1180 (s), 934 (s), 817 (s),
726 (w), 675 (s), 615 (s), 568 (s), 501 (s), 462 (s). Anal. Calcd
C12H27SiSBrFe (desolvated): C, 39.2; H, 7.4; Br, 21.8. Found: C,
39.7; H, 7.3; Br, 21.8. A quench with DCl/CD3OD afforded thiolate/
1
C6H6 ) 1.0:1.0 by H NMR spectroscopic analysis.
2.12. [Fe(µ-I)(µ-SSitBu3)]14(C6H6)14 (6-FeI). A 50 mL flask
was charged with NaSSitBu3(THF)1.56 (0.301 g, 0.820 mmol), FeI2-
(THF)2 (0.371 g, 0.817 mmol), and THF (20 mL). The mixture
was stirred at 23 °C for 24 h, and the solvent was removed to yield
a yellow solid. The solid was heated at 117 °C under vacuum for
5 h then extracted into benzene and filtered. After slow evaporation
of the solvent (7 days), dark yellow crystals formed (0.064 g, 16%).
1H NMR (C6D6): insoluble. IR (Nujol Mull, NaCl, cm-1): 1190
(s), 1180 (s), 1012 (s), 936 (s), 860 (w), 815 (s), 726 (w), 675 (m),
615 (s), 566 (s), 492 (s), 462 (s). Anal. Calcd C12H27SiSIFe
(desolvated): C, 34.8; H, 6.6; I, 30.6. Found: C, 28.8; H, 5.2; I,
30.3. A quench with DCl/CD3OD afforded thiolate/C6H6 ) 1.0:
1
0.9 by H NMR spectroscopic analysis.
2.13. [Co(µ-Cl)(µ-SSitBu3)]12(C6H6)6 (5-CoCl). A 100 mL flask
was charged with CoCl2 (0.347 g, 2.673 mmol), NaSSitBu3(THF)1.66
(1.003 g, 2.680 mmol), and THF (60 mL). The mixture was stirred
at room temperature for 12 h, and the solvent was removed under
vacuum. The blue solid was heated at 81°C under vacuum for 1.5
h to give a green solid that was extracted into benzene and filtered.
Slow evaporation of the benzene yielded the product as green
microcrystals (0.206 g, 21%). 1H NMR (C6D6): δ 9.11 (υ1/2 ≈ 50
Hz). IR (Nujol Mull): 1189 (m), 1013 (s), 934 (m), 816 (s), 675
(m), 614 (s), 563 (s) cm-1. Anal. Calcd for C12H27SiSClCo
(desolvated): C, 44.2; H, 8.4; Cl, 10.9. Found: C, 44.2; H, 8.4; Cl,
10.9. A quench with DCl/CD3OD afforded thiolate/benzene )
1
1.0:0.5 by H NMR spectroscopic analysis.
2.14. [(tBu3SiS)Ni]2(µ-SSitBu3)2 (7). A 25 mL flask was charged
with NiBr2(THF)2 (0.100 g, 0.276 mmol), NaSSitBu3(THF)1.54
(0.200 g, 0.547 mmol), and THF (20 mL). The mixture was stirred
at 23 °C for 24 h, and the solvent was removed under vacuum.
The purple solid was extracted with diethyl ether and filtered. The
solution was concentrated and crystallized at -78 °C to yield the
4.2. [(tBu3SiS)Fe]2(µ-SSitBu3)2 (12). An orange block (0.6 × 0.5
× 0.4 mm3) from pentane was used. Data collection on a Siemens
P4 diffractometer gave 4008 of the 4940 unique data points
(81.13%, Rint ) 0.0654) where |Fo| > 2σ|Fo|. Refinement utilized
1
product as purple microcrystals (0.091 g, 63%). H NMR (C6D6):
δ 1.31 (54 H, -SSi(C(CH3)3)3), 1.35 (54 H-SSi(C(CH3)3)3). 13C-
{1H} NMR (C6D6): δ 25.84 (C(CH3)3), 26.41 (C(CH3)3), 31.97
(C(CH3)3), 32.25 (C(CH3)3). IR (Nujol Mull): 1182 (m), 1012 (m),
934 (m), 815 (s), 611 (s), 563 (s) cm-1. Anal. Calcd for C24H54-
Si2S2Ni: C, 55.2; H, 10.5. Found: C, 54.9; H, 10.9. Mw (Signer
isopiestic method; solvent ) pentane, standard ) ferrocene)
calcd: 1044, found: 1030 ( 60 (3 trials).
2
SHELXTL PLUS and w-1 ) σ2(Fo ) + (0.1072p)2 + 4.0995p,
2
where p ) (Fo + 2Fc2)/3).
4.3. (tBu3SiS)2Fe(THF)2 (1-(THF)2). A yellow block (0.4 ×
0.3 × 0.2 mm3) from pentane was selected. Data collection on a
Siemens P4 diffractometer revealed 5733 of the 6047 unique data
2.15. [Ni(µ-Br)(µ-SSitBu3)]12(C6H14)n (5-NiBr). A 25 mL flask
was charged with NiBr2(THF)2 (0.150 g, 0.414 mmol), NaSSitBu3-
(THF)1.54 (0.150 g, 0.410 mmol), and THF (20 mL). The mixture
was stirred at 23 °C for 16 h, and the solvent was removed under
vacuum. The green solid was heated at 88 °C under vacuum for 3
h, yielding a red-purple solid that was washed with diethyl ether
to remove 7 via disproportionation. The resulting green solid was
extracted with hexane and filtered. Slow evaporation of the hexane
over 12 h yielded the product as a microcrystalline red solid (0.026
g, 17% based on desolvated 5-NiBr). 1H NMR (C6D12): insoluble.
points (94.81%, Rint ) 0.0271) where |Fo| > 2σ|Fo|. Refinement
2
utilized SHELXTL PLUS and w-1 ) σ2(Fo ) + (0.0855p)2
+
2
1.4794p, where p ) (Fo + 2Fc2)/3).
4.4. [Br2Fe](µ-SSitBu3)2[FeBr(THF)]Na(THF)4 (3-Br). A yel-
low block (0.4 × 0.3 × 0.2 mm3) obtained from THF was used.
Data collected on a Siemens SMART CCD Area Detector system
were processed with the Bruker SAINT program (17 443 reflections,
6841 were symmetry independent, Rint ) 0.0542). The data were
corrected for absorption with SADABS and refined using SHELX-
Inorganic Chemistry, Vol. 45, No. 2, 2006 625

Sydora et al.
TL. Both of the tBu3Si groups exhibited rotational disorder around
the Si-S axis, which was modeled by refining the occupancy (50%)
of two sets of tBu3Si groups. The model did not refine well, but no
better one was found.
equivalent cavities centered at (0, 0, 0) and (1/2, 1/2, 0) with
volumes of 2052 Å3 (cell volume ) 13 349(6) Å3) were each
occupied by 805 electrons. Since the crystallization was carried
out in benzene solution, the asymmetric unit could contain, on
average, 4.8 benzene molecules. The final model consisted of the
ordered part only, without the disordered solvent contribution, and
was refined against new data provided with the fcf file.
4.5. cis-[(THF)IFe]2(µ-SSitBu3)2 (4). A yellow block (0.2 × 0.2
× 0.15 mm3) was obtained from C6H6. Data collected on a Siemens
SMART CCD Area Detector system were processed with the
Bruker SAINT+ program (19 236 reflections, 6142 were symmetry
independent, Rint ) 0.0263). Refinement utilized SHELXTL.
4.6. [Fe(µ-Cl)(µ-SSitBu3)]12(C6H6)7 (5-FeCl). A yellow crystal
(0.2 × 0.15 × 0.05 mm3) was obtained from benzene. Data
collected on a Siemens SMART CCD Area Detector system were
processed with the Bruker SMART program (46 598 reflections,
9453 were symmetry independent, Rint ) 0.0726). The data were
corrected for absorption with SADABS and refined using SHELX-
TL. All non-hydrogen atoms were refined with anisotropic dis-
placement parameters, except for half a benzene ring lying on a
special position, and hydrogen atoms were included at calculated
positions. Two and a half benzene molecules were found in the
asymmetric unit. Data refinement indicated partial occupancy of
the benzene sites, but independent refinement of the benzene
4.9. [Ni(µ-Br)(µ-SSitBu3)]12(C6H14)n (5-NiBr). A red block
crystal (0.5 × 0.6 × 0.6 mm3) was obtained from pentane. Data
collected on a Siemens SMART CCD Area Detector system were
processed with the Bruker SAINT program (9646 reflections, 3730
were symmetry independent, Rint ) 0.1259). The data were
corrected for absorption with SADABS and refined using SHELX-
TL. The carbon atoms were not refined anisotropically due to
disorder that could not be incorporated into the model due to poor
data quality. Constraints were applied to the geometries of the
t
tBu3Si fragments on each Bu3SiS ligand such that chemically
equivalent inter- and intra-atomic distances were constrained to
equal the same least-squares variables (e.g., all d(SiC) are equiva-
lent; all d(CC) are equivalent, etc.). The unit cell had two cavities
containing a large amount of severely disordered solvent molecules.
The difference Fourier map in this region showed a number of
peaks, but no consistent model could be assembled. This part of
the structure was modeled using the SQUEEZE procedure of the
PLATON program.19 Two cavities (per cell) of 3012 Å3 (cell
volume ) 14 597 Å3) were each occupied by 464 electrons. The
final model consisted of the ordered part only, without the
disordered solvent contribution, and was refined against new data
provided with the fcf file.
4.10. [Fe(µ-I)(µ-SSitBu3)]14(C6H6)14 (6-FeI). A yellow crystal
(0.5 × 0.4 × 0.2 mm3) was obtained from benzene. Data collected
on a Siemens SMART CCD Area Detector system were processed
with the Bruker SAINT+ program (104 844 reflections, 10 375
were symmetry independent, Rint ) 0.0961). Refinement utilized
SHELXTL but the carbon and silicon atoms were refined isotro-
pically due to disorder that could not be incorporated into the model
due to poor data quality. The unit cell had cavities containing a
large amount of severely disordered benzene molecules. One
ordered benzene was located in the center of the ring, a partially
disordered benzene was located outside the ring, and several
completely disordered benzenes could not be modeled. This part
of the structure was modeled using the SQUEEZE procedure of
the PLATON program.19 The final model consisted of the ordered
part only, without the disordered solvent contribution, and was
refined against new data provided with the fcf file.
t
occupancies did not yield a Bu3SiS/C6H6 ratio (1:0.2) consistent
with the experimentally determined value of 1:0.6. The occupancy
values were set to 75% for the two benzene molecules and 50%
for the half benzene molecule to correspond exactly to the
experimental value.
4.7. [Co(µ-Cl)(µ-SSitBu3)]12(C6H6)6 (5-CoCl). A green block
crystal (0.3 × 0.2 × 0.15 mm3) was obtained from benzene. Data
collected on a Siemens SMART CCD Area Detector system were
processed with the Bruker SAINT program (48 983 reflections, 8000
were symmetry independent, Rint ) 0.1040). The data were
corrected for absorption with SADABS and refined using SHELX-
TL. Two and a half benzene molecules were found in the
asymmetric unit. Data refinement indicated partial occupancy of
the benzene sites. The benzene occupancies were refined indepen-
dently, with the two benzene occupancies refining to 68 and 71%,
while the occupancy of the half-benzene refined to 46%. These
values gave a tBu3SiS/C6H6 ratio (1:0.54) very close to that obtained
by experimental methods (1:0.5), so the occupancy values were
set to 62.5% for the two benzene molecules and 50% for the half
benzene molecule to correspond exactly to the experimental value.
4.8. [Fe(µ-Br)(µ-SSitBu3)]12(C6H6)12 (5-FeBr). A yellow crystal
(0.3 × 0.2 × 0.05 mm3) was obtained from benzene. Data collected
on a Siemens SMART CCD Area Detector system were processed
with the Bruker SAINT+ program (17 324 reflections, 5190 were
symmetry independent, Rint ) 0.0976). The data were corrected
for absorption with SADABS and refined using SHELXTL. The
carbon atoms were not refined anisotropically due to disorder that
could not be incorporated into the model due to poor data quality.
The unit cell had two cavities containing a large amount of severely
disordered benzene molecules. The difference Fourier map in this
region showed a number of peaks, but no consistent model could
be assembled. This part of the structure was modeled using the
SQUEEZE procedure of the PLATON program.19 Two symmetry-
Acknowledgment. We thank the National Science Foun-
dation (CHE-9528914 (P.T.W.)) and Cornell University for
financial support.
Supporting Information Available: CIF files for 12, 1-(THF)2,
3-Br, and 4. This material is available free of charge via the
Internet at http://pubs.acs.org.
IC051289U
626 Inorganic Chemistry, Vol. 45, No. 2, 2006